Optimize Deep Learning Projects on Upwork

I need a complete, end-to-end pipeline that starts with a screenshot of a live trading terminal, pulls every ticker, quote, or indicator visible on the image, and ends with a single buy-or-sell probability score that consistently reaches 90 %+ confidence.

Scope
• Screenshot ingestion and parsing: an OCR/visual-recognition routine that can handle typical broker or data screens, isolate each symbol, price, volume and time stamp, and structure it as tabular data.
• Feature engineering: enrich the extracted snapshot with any historical data you deem necessary (you may tap your own CSV archives or free APIs), then create derived variables such as intraday return, volatility, or technical signals.
• Probabilistic modeling: build and train a machine-learning model—choice of algorithm is open, as long as it is explainable and achieves the target accuracy on held-out data. Neural networks, tree ensembles or hybrid methods are all acceptable if they outperform simpler baselines.
• Scoring interface: a lightweight CLI or web endpoint that accepts a new screenshot and returns a JSON payload containing “BUY”, “SELL” and the associated probability.
• Documentation & reproducibility: clean, commented Python code (TensorFlow, PyTorch, scikit-learn or similar), requirements.txt, and a short read-me describing set-up and retraining steps.

Acceptance criteria
1. End-to-end demo with two unseen screenshots producing valid scores.
2. Back-test report showing at least 90 % directional accuracy across 1,000+ historical samples.
3. Code passes linting and runs on a fresh environment with the provided instructions.

If parts of the stack—OCR, data augmentation, or model selection—require iterations, please outline your milestone plan so we can agree on checkpoints and validation tests before moving forward.

Skills: Python, Matlab and Mathematica, Software Architecture, Machine Learning (ML), C++ Programming, Data Science, Neural Networks, Deep Learning, Livestock

Your project specification is already highly detailed. Below is a refined and more professionally structured version with the new dual-layer system architecture added properly:
Training Mode
Prediction Mode
This structure is cleaner, more scalable, and closer to enterprise AI system design.
APP DEVELOPMENT SPECIFICATION
APPLICATION NAME
Amir B From Malaysia
FCPO Shariah Compliant Market Only
============================================================
PLATFORM & DEPLOYMENT
Operating System
Windows Platform
Application Interface
Command Prompt Interface (CMD)
Deployment Format
Standalone Windows .EXE Application
Execution Environment
Fully Offline Execution
Lightweight System Architecture
High-Speed Processing Engine
Optimized CPU & RAM Utilization
Stable for Continuous Daily FCPO Analysis
Fast Startup & Low-Latency Prediction Engine
Security & Licensing
Encrypted License Key Authentication System
License Options
Lifetime License
3-Year Expiry License
License Requirement
Application cannot execute without valid license authentication
============================================================
SYSTEM OBJECTIVE
The application must analyze historical FCPO OHLC Excel datasets and generate:
Next-Day Market Direction Prediction
Same-Day Market Direction Prediction using today Open price before market close
The AI framework must integrate:
Machine Learning
Statistical Modeling
Volatility Modeling
Deep Learning
Market Structure Analysis
Ensemble Optimization
Adaptive Market Learning
============================================================
MARKET PREDICTION LOGIC
BUY
Expected Close Price > Open Price
SELL
Expected Close Price < Open Price
Optional: NO TRADE
Triggered when:
prediction confidence is weak
market structure is unstable
AI consensus is conflicting
volatility condition is abnormal
============================================================
DATA SOURCE & EXCEL INTEGRATION
The application must support manual Excel integration.
Supported Excel Columns
Date
Open
High
Low
Close
User Functions
Paste Excel file directory manually
Update Excel data anytime
Save latest candle data manually
Instantly rerun analysis
Automatically retrain after new data entry
Example Path
D:\FCPO\data.xlsx
============================================================
TRAINING MODE ARCHITECTURE
The application must separate:
AI Training Mode
Prediction Mode
This allows flexible combinations between:
full retraining
saved model execution
next-day prediction
same-day prediction
============================================================
TRAINING MODES
1. RUN USING LAST MACHINE LEARNING
Purpose: Use the latest saved trained AI models for faster execution.
Functions:
Load latest trained model state
Load latest optimized ensemble weighting
Perform incremental retraining on newest candle data only
Fast operational analysis mode
Optimized for daily usage
Advantages:
Faster execution
Lower CPU usage
Faster prediction generation
Suitable for continuous daily operation
============================================================
2. RE-RUN MACHINE LEARNING FROM FIRST DATA
Purpose: Perform complete AI rebuilding from earliest historical dataset.
Functions:
Retrain all models from beginning
Rebuild ensemble structure
Recalculate feature importance
Perform deep optimization cycle
Execute full walk-forward backtesting
Recalibrate confidence weighting
Advantages:
Maximum optimization quality
Full AI recalibration
Better long-term adaptation
More stable ensemble rebuilding
============================================================
PREDICTION MODES
1. RUN NEXT-DAY MARKET PREDICTION
Purpose: Predict next FCPO trading day direction using completed OHLC historical candles.
Prediction Output:
BUY
SELL
NO TRADE
Analysis Components:
historical market behavior
volatility structure
momentum continuation
market regime analysis
trend continuation probability
ensemble confidence weighting
============================================================
2. RUN TODAY MARKET PREDICTION
Purpose: Predict whether the current FCPO trading day is likely to close bullish or bearish BEFORE market close.
The prediction must use today Open price together with historical FCPO market behavior and live market structure analysis.
============================================================
SAME-DAY AI INPUT ENGINE
For stronger same-day prediction accuracy, the AI engine must analyze:
Today Open Price
Previous Day OHLC
Previous Volatility Structure
Previous Momentum Structure
Current Market Regime
Historical FCPO Market Behavior
Historical Trend Continuation Patterns
AI Ensemble Probability Distribution
Optional Additional Inputs
Intraday Price Movement
Intraday Volatility Expansion
Early Session Momentum Shift
============================================================
SAME-DAY MARKET PREDICTION LOGIC
BUY
Expected Today Close > Today Open
SELL
Expected Today Close < Today Open
Optional: NO TRADE
Triggered when:
intraday structure becomes unstable
AI confidence weakens
volatility becomes abnormal
ensemble signals conflict
============================================================
LIVE MARKET ANALYSIS ENGINE
The system must dynamically evaluate:
bullish vs bearish pressure
trend continuation probability
volatility expansion probability
market structure transition
historical behavioral similarity
AI ensemble confidence
regime transition probability
momentum persistence strength
Prediction confidence must improve dynamically as more market data becomes available during trading hours.
============================================================
AI & MATHEMATICAL ENGINE
The application must initially test ALL mathematical models during historical optimization and backtesting before determining final ensemble weighting.
CORE AI ENGINE
XGBoost
LightGBM
CatBoost
SVM
VOLATILITY ENGINE
GARCH
EGARCH
Heston Model
Stochastic Volatility Model
MARKET STRUCTURE ENGINE
Structural VAR
VECM
Hidden Markov Model (HMM)
TREND FILTER ENGINE
Kalman Filter
DEEP LEARNING ENGINE
LSTM
GRU
SUPPORTING MODELS
Random Forest
Bayesian Network
ARIMA
SARIMA
Monte Carlo Simulation
FINAL AI ENGINE
Stacking Ensemble Model
============================================================
AI SYSTEM DESIGN REQUIREMENTS
The application must NOT remove any mathematical models before full historical backtesting and optimization are completed.
System Requirements
test all models
test all model combinations
compare historical accuracy
compare prediction stability
compare retraining efficiency
compare overfitting risk
compare market adaptability
compare long-term consistency
compare regime sensitivity
compare confidence calibration
Post-Optimization Logic
After optimization:
weaker models may receive lower weighting
unstable models may be disabled
stronger models may receive higher ensemble priority
Primary Objectives
highest realistic prediction accuracy
stable BUY/SELL prediction
adaptive market learning
smooth retraining workflow
fast analysis execution
sustainable long-term stability
minimal overfitting risk
strong confidence calibration
stable live-market prediction performance
============================================================
SMART AUTO-RETRAINING SYSTEM
The application must automatically retrain whenever new Excel candle data is detected.
When new candle data is added:
System Actions
Detect newly added rows automatically
Update:
market structure
volatility regime
trend direction
AI feature layers
historical sequence learning
ensemble confidence
momentum state
market condition classification
Retrain AI models automatically
Reoptimize:
ensemble weighting
feature importance
model confidence
prediction stability
signal reliability
model prioritization
Generate latest prediction automatically
============================================================
USER WORKFLOW
STEP 1
Open Application
STEP 2
Enter License Key
STEP 3
Paste Excel File Directory
STEP 4
Select Training Mode
RUN USING LAST MACHINE LEARNING
RE-RUN MACHINE LEARNING FROM FIRST DATA
STEP 5
Select Prediction Mode
RUN NEXT-DAY MARKET PREDICTION
RUN TODAY MARKET PREDICTION
STEP 6
Click RUN ANALYSIS
STEP 7
Application automatically:
load Excel dataset
validate market data
update AI learning
retrain models
optimize ensemble
generate prediction
display final market signal
============================================================
SYSTEM ARCHITECTURE
Frontend
Command Prompt Interface
Backend
Python AI Engine
Data Source
Excel OHLC Dataset
AI Framework
Hybrid Machine Learning + Statistical Models
Training System
Smart Automatic Retraining Engine
Prediction Output
BUY / SELL / NO TRADE
Deployment
Standalone Windows .EXE Application
============================================================
FINAL DELIVERY REQUIREMENT
The entire application must be fully completed, including:
AI engine development
smart retraining architecture
optimization workflow
historical backtesting engine
Windows .EXE packaging
encrypted license key system
Excel integration
automatic retraining workflow
prediction logging system
user operation manual
Final User Workflow
Download application
Enter license key
Link Excel dataset
Select training mode
Select prediction mode
Run analysis
============================================================

Skills: Pine Script, Trading, AI Development

Hi - I am already quite skilled in AI - but I am looking for an expert in LLM fine tuning - specifically for models such as Qwen or Mistral without the weighting restrictions on it. I need to train an LLM and fine tune it with 100,000+ chat conversations and make the LLM unrestricted with requests.

You must be an expert in LLM fine tuning - specifically in this area. I can fine tune myself - so please do not try to sell me over-priced products or unnecessary things or go on Hugging Face and just submit a random model.

The problem is that if you remove all the weights, it can easily hallucinate which is what we do not want.

We are looking for a remote monitoring professional to cover 8PM-8AM.
We provide EEG testing solutions.

The right candidate will have experience with EEG and related software.

We are seeking an OpenClaw expert to enhance our custom AI model. The ideal candidate will have experience in AI model development and optimization, with a focus on improving performance and efficiency. Responsibilities include analyzing current AI systems, identifying areas for improvement, and implementing OpenClaw solutions to enhance functionality and speed. The role requires collaboration with our team to integrate new features and ensure seamless operation.

Lacrosse Rebounder AI Tracking App
Prepared by: Raza Date: May 9, 2026
Overview
An on-device iOS computer vision and audio fusion system that automatically detects, counts,
and classifies lacrosse rebounder reps in real time. Core logic delivered as a clean module with
endpoints for your developer to integrate into your existing app. No server dependency. Works
fully offline.
Phase 1: Proof of Concept $500
Before full development begins, I will process your existing front-view footage and deliver a
demo clip so you can see the tracking working before committing to the full build.
Deliverables:
• Processed video clip of ~10 reps from your head-on footage
• Pose skeleton overlay (full body landmarks)
• Stick tracking overlay
• Ball tracking overlay
• Audio-triggered rep detection (ball impact thud as trigger)
• Left/right handedness classification
• Catch / no catch classification
• Investor-ready demo clip
Timeline: 3 to 4 business days from footage received and payment confirmed.
If you are satisfied, we move to the full build. If not, we stop no further commitment required.
Full Build $5,500
One milestone. Full on-device core logic delivered via TestFlight for testing, with clean
endpoints ready for your developer to integrate into your existing app.
Deliverables:
• On-device iOS core module, no server dependency, works offline
• Live camera feed with real-time pose, stick, and ball tracking
• Audio fusion rep trigger (ball impact sound detection)
• Per-rep classification: left/right handedness, catch/no catch, throw type
• Additional throw types: top hand, behind the back, quick-stick
• Throw quality and crispness grading
• Basic form analysis
• Rep counter and session tracking
• Voice-guided camera positioning
• Session history and rep log
• Clean endpoints exposed for your developer to hook into your existing app
• TestFlight delivery for testing and validation
Note on Integration: Your developer will be responsible for the frontend integration and linking
up the endpoints. If you can share access to the developer account I will take a quick look at the
existing code to make sure we are fully aligned on the integration points before starting.
Timeline: 6 to 8 weeks from Phase 1 approval.
Next Steps
1. Confirm this proposal
2. Share your front-view finalized footage
3. Demo clip delivered in 3 to 4 days
4. Full build kicks off upon your approval
Looking forward to building this.
Best, Raza

Complete Lottery Prediction and Betting Automation System (Focused on Loterías y Apuestas del Estado - Spain)
2. System Features
2.1. Historical Data Collection and Update
The system must automatically download complete historical results (drawn numbers, draw dates, prize breakdowns by category, accumulated jackpots) from the first draw of each lottery, directly from loteriasyapuestas.es or reliable associated sources.
Specific sources:

Euromillones: loteriasyapuestas.es/es/euromillones (since Feb 13, 2004)
La Primitiva: loteriasyapuestas.es/es/la-primitiva (since Oct 17, 1985 – modern version)
El Gordo de la Primitiva: loteriasyapuestas.es/es/gordo-primitiva (since Oct 31, 1993)

Updates automatic at exactly 00:02 the day after each draw, using ethical scraping (BeautifulSoup/Scrapy) with proper user-agent headers to mimic human behavior.
Store data in PostgreSQL (structured) or MongoDB (flexible), including all prize categories to enable ROI calculations and backtesting.
2.2. Number Prediction
Generate predictions for Euromillones, La Primitiva and/or El Gordo simultaneously using explicit advanced AI models:

Machine Learning ensembles (Random Forests) for frequency/statistical patterns
Deep Learning RNN/LSTM for time-series analysis and sequential trends
Neuro-Symbolic AI to combine logical rules with neural learning
Evolutionary/Genetic Algorithms to evolve optimal combinations
Bayesian Networks for conditional probabilities between numbers

Analyze full history (numbers, frequencies, temporal patterns, prizes, jackpots) and incorporate:

Hot/cold number analysis
Wheeling systems for guaranteed coverage of partial prizes
Monte Carlo simulations to estimate expected value (EV) for high jackpots

Generate blocks of 10 bets at a time using itertools for diversity.
User selects total bets (predefined: 10, 20, 50, 100, 500, 1,000, 5,000 or custom), with AI suggestions based on current jackpot (daily scraped), historical odds and bankroll.
After each real draw: compare results, report additional bets needed to hit jackpot (what-if simulations with genetic algorithms), and detail prizes won (hypothetical or actual).
Full backtesting: simulate from first historical draw, calculate bets required to cover winner (dynamic, no fixed limit – supports thousands/millions hypothetical bets), using Evolutionary Algorithms; report hit rates, average hypothetical ROI, patterns via LSTM/Bayesian.
2.3. Flexible Draw Days
Allow manual/automatic changes to draw days (official changes/holidays), stored dynamically in DB.
Interactive calendar (FullCalendar) with email/SMS alerts and real-time scraping for schedule updates.
2.4. Bet Cost Calculation
Auto-calculate total cost based on user-selected bets (blocks of 10+), fixed prices:

Euromillones: €2.50
La Primitiva: €1.00
El Gordo: €1.50

Dynamic AI suggestions: recommend bet quantity based on backtesting, current jackpot and balance, using Monte Carlo for EV (prioritize positive EV scenarios).
2.5. Login and Bet Execution Automation
Never store login/password (not even encrypted). Prompt user for credentials (email/NIF/NIE + password) every login, using secure masked fields.
Login process: Selenium (headless Chrome/Firefox), access https://juegos.loteriasyapuestas.es/jugar/login, fill form, submit, verify success (post-login elements), handle sessions/cookies.
Bet execution: navigate to lottery page, generate AI numbers, add exact user-requested bets to cart (batches for high volumes), support rules of each lottery.
Finalization: show full summary (lottery, bets, cost, numbers); require explicit user confirmation before submitting (“Confirm and Submit Bets”).
Payment handling: scrape Lotobolsa balance post-login; if insufficient, pause, notify user exact top-up amount via registered bank account (BBVA/CaixaBank transfer, min €1), wait for manual confirmation or recheck; proceed only on sufficient balance + user confirmation. Capture receipt, notify user, log details (no credentials).
Security: random delays, CAPTCHA handling (user intervention), geoblock alerts, expired session relogin, respect site limits. All via Selenium (no public API).
2.6. Simulations and Reports
Historical backtesting: complete simulations from first draw, report per draw: bets needed to cover jackpot (dynamic via Evolutionary Algorithms), hypothetical prizes, metrics (hit rate, ROI, patterns via LSTM/Bayesian).
Post-draw reports: actual matches, minimum bets to win, next-cycle suggestions.
User options: custom simulations, bet quantity definition, PDF/Excel export with charts.
2.7. User Interface

Prediction screen: numbers in blocks of 10, probabilities, wheeling suggestions, bet adjustment
Betting screen: lottery selection, bet input, cost & EV simulations, start button
Login/Config: credential fields (each use), login test, bankroll
Confirmation screen: summary + mandatory user confirmation
Agenda: interactive calendar with alerts
Backtesting: graphical reports (Matplotlib/Plotly) – frequencies, performance, ROI
Responsive design, intuitive dashboards (heatmaps, multi-lottery support).

3. Detailed Implementation
3.1. Data Structure: DB schemas for lotteries (rules/ranges/draw days), full history (indexed), user settings (no credentials).
3.2. Draw Day Changes: internal API with official validation & calendar sync.
3.3. Cost Calculation: dynamic Python functions + jackpot scraping & EV.
3.4. Prediction: train RNN/LSTM (PyTorch/TensorFlow), combinatorial generation (itertools), Genetic optimization.
3.5. Backtesting: parallel scripts (multiprocessing), no bet limits, model cross-validation.
3.6. Automation: as in 2.5, with execution logs and real-time UI feedback.
4. Conclusion
System developed for exclusive personal use, integrating data collection, multi-model AI optimization, supervised secure automation with user confirmations, and comprehensive reporting.

Skills: Web Scraping, Data Analysis, Deep Learning, Selenium

We are looking for an AI Engineer with experience in Machine Learning, Deep Learning, and Computer Vision to build intelligent and scalable AI solutions. The ideal candidate should have strong skills in Python, AI model development, and automation workflows. Experience with TensorFlow, PyTorch, OpenCV, and real world AI applications is preferred. This role is ideal for someone passionate about developing high performance AI systems and innovative solutions.

Brain disorder detection using Deep Learning
Uses MRI brain scan images
Detects multiple brain diseases automatically
Early disease detection and diagnosis
Multi-class classification system
Diseases detected:
Glioma tumor
Pituitary tumor
Meningioma tumor
Alzheimer’s disease
Healthy brain
Uses Convolutional Neural Networks (CNN)
Uses Transfer Learning techniques
Models used:
CNN
VGG16
ResNet50
Dataset collected from Kaggle MRI datasets
Combined brain tumor and Alzheimer datasets
1000 images used for training
250 images used for testing
Image preprocessing and normalization
Pixel values converted from 0–255 to 0–1
Feature extraction using convolution layers
Pooling layers reduce computation
Fully connected layers perform classification
Softmax activation used for multi-class prediction
Hyperparameter tuning performed
Parameters tuned:
Learning rate
Epochs
Batch size
Filter size
Number of filters
Best performance achieved with low learning rate and more epochs
Accuracy achieved above 90%
Performance metrics used:
Accuracy
Precision
Recall
F1-score
Confusion matrix
Technologies used:
Python
TensorFlow
Keras
OpenCV
NumPy
Pandas
Matplotlib
Scikit-learn
Platform used:
Google Colab
Anaconda
Model deployment using Hugging Face and Streamlit
Upload MRI image to predict disease
Fast and automated diagnosis system
Reduces manual work for doctors
Helps in early treatment planning
Non-invasive medical diagnosis method
Future scope:
Add more diseases
Improve accuracy
Use larger datasets
Clinical healthcare implementation

Skills: Python, Statistics, Machine Learning (ML), Data Mining, OpenCV, Keras, Deep Learning, Convolutional Neural Network

Job Title: AI Tools Specialist (ChatGPT, Claude, Gemini)

We are looking for a skilled freelancer who has experience working with AI tools such as ChatGPT, Claude, and Gemini.

Responsibilities:
Use AI tools to generate high-quality content and responses
Assist with data entry, research, and writing tasks
Create accurate, well-structured outputs based on instructions
Support daily operational tasks using AI tools
Requirements:
Experience with ChatGPT, Claude, or Gemini
Good English communication skills
Ability to follow instructions carefully
Detail-oriented and fast learner
Work Type:
Remote / Freelance
Part-time or full-time available

We are seeking a skilled Python engineer with strong experience in computer vision to develop and enhance our image processing capabilities. The ideal candidate will have a solid understanding of point cloud processing and be able to implement efficient algorithms for image processing tasks.

I'm seeking a tutor for computational image processing and data-driven methods (machine/deep learning), specifically for my biomedical engineering coursework.

I'm looking for assistance in:
- Image segmentation
- Feature extraction
- Image registration
- Binary images
- Image transforms

The exam is mathematical and theoretical.

Ideal skills and experience:
- Expertise in computational image processing
- Proficiency in data analysis, visualization, and statistical methods
- Excellent teaching and communication skills

Looking forward to your expertise!

Skills: Machine Learning (ML), Mathematics, Data Mining, Neural Networks, Image Processing, Data Visualization, Data Analysis, Computer Vision, Deep Learning, Image Analysis

🤖 AI/ML Engineer (Remote – US Time Zone Friendly)

We are hiring an AI/ML Engineer to design, build, and deploy production-grade machine learning systems and LLM-powered applications for fast-growing international digital products.

We’re looking for engineers who can work like a product builder — not just a model trainer.

💡 What You’ll Work On

You will help build and scale:

LLM-powered applications (AI assistants, automation tools)
Machine learning models in production environments
NLP-based features and intelligent systems
Scalable AI pipelines from data → training → deployment
💻 Key Responsibilities
Design and train machine learning / deep learning models
Build and deploy LLM/NLP solutions into production systems
Develop APIs for ML model integration (FastAPI / Flask)
Work with large datasets, feature engineering, and optimization
Improve model accuracy, performance, and scalability
Collaborate with backend, product, and engineering teams
🧠 Required Skills
Strong Python development experience
Hands-on experience with PyTorch / TensorFlow / Scikit-learn
Strong knowledge of NLP, LLMs, or predictive modeling
Experience deploying ML models to production
Familiarity with REST APIs and cloud environments (AWS / GCP / Azure)
Strong problem-solving and system design mindset
⭐ Preferred (Nice to Have)
Experience with OpenAI / LLM APIs (GPT, embeddings, agents)
MLOps tools (Docker, CI/CD, MLflow)
Vector databases (Pinecone, FAISS, Weaviate)
Experience working in US startups or SaaS products
Strong understanding of scalable architecture
🌍 Role Details

Remote (US time zone overlap preferred)
Long-term opportunity
Startup-style, fast-moving engineering environment

🚀 Why This Role
Work on real AI products used internationally
High-impact engineering ownership (not just tasks)
Fast decision-making, minimal bureaucracy
Modern AI stack (LLMs, automation, scalable ML systems)
Remote-first collaboration with global teams

We are looking to develop an AI-based solution capable of determining the authenticity of ID document images captured from mobile devices.

The system should analyze each image using five distinct filters. Each filter will answer a specific question and return a confidence percentage. Detailed information about these filters, as well as the expected output format, can be found in the attached document.

The solution must be delivered as an API service. The API should accept one or multiple images as input and return a structured response matching the specifications outlined in the document.

This solution will be integrated into a KYC application, so accuracy and reliability are critical. We are specifically looking for someone with proven experience in similar projects—ideally involving computer vision, image forensics, or document verification. If you do not have relevant experience in these areas, we kindly ask that you do not submit a proposal.

Our preference is to build this solution using proprietary code, minimizing reliance on third-party services. The goal is to maintain control over costs and offer this capability to our clients at a competitive price. If GPU processing is required, we currently use RunPod, so experience with GPU-based workloads or similar infrastructure would be valuable.

To confirm that you have read and understood these requirements, please begin your proposal with the words: ADAMO ID

1. Smart CT Scan Analyzer

An AI system that analyzes lung CT scan images and detects possible lung cancer signs early.

Main Features
Upload CT scan or X-ray images
AI detects lung nodules
Cancer risk prediction
Heatmap highlighting suspicious areas
Doctor dashboard
Patient history tracking
PDF medical report generation
Technologies
Frontend: Vue.js / React
Backend: Node.js or Python Flask
AI Model: TensorFlow / PyTorch
Database: MySQL / PostgreSQL
Image Processing: OpenCV
AI Models
CNN (Convolutional Neural Network)
ResNet50
YOLO for object detection
Users
Doctors
Radiologists
Hospitals
Patients
2. AI + IoT Lung Monitoring System

A smart healthcare platform connected to wearable devices.

Features
Real-time breathing monitoring
Oxygen level tracking
AI predicts lung disease risk
Emergency alerts
Mobile app notifications
Patient monitoring dashboard
Hardware
ESP32
Pulse Oximeter Sensor
Temperature Sensor
AI Functions
Predict worsening symptoms
Detect abnormal breathing patterns
Risk scoring
3. Cloud-Based Lung Cancer Prediction Platform

A hospital management + AI diagnosis platform.

Modules
Patient Management
Appointment Booking
AI Diagnosis
Lab Result Upload
Doctor Recommendation
Treatment Tracking
Special AI Features
Predict cancer stage
Suggest possible treatments
Survival rate estimation
Drug recommendation support
4. AI Chat Assistant for Lung Cancer Patients

An intelligent assistant helping patients understand symptoms and treatment.

Features
Symptom checker
AI chatbot
Medicine reminders
Treatment education
Doctor communication
Extra
Voice support
Multiple languages
WhatsApp integration
5. Advanced Real-World System (Best for Final Year Project)
System Name

LungAI Pro

Full Modules
Patient Side
Registration/Login
Upload scans
View AI results
Book appointments
Payment system
Doctor Side
Review AI predictions
Add notes
Manage patients
Generate reports
Admin Side
Manage hospitals
Manage doctors
Analytics dashboard
AI training management
AI Workflow
CT Scan Image
↓
Preprocessing
↓
AI Model Analysis
↓
Tumor Detection
↓
Cancer Classification
↓
Generate Diagnostic Report
Database Tables Idea
patients
id
name
email
phone
gender
date_of_birth
scans
id
patient_id
scan_image
scan_type
uploaded_at
ai_results
id
scan_id
prediction
confidence_score
cancer_stage
doctors
id
name
specialization
hospital
appointments
id
patient_id
doctor_id
appointment_date
status
Advanced AI Features
Possible Detection Types
Benign vs Malignant tumor
Lung nodule segmentation
Cancer stage classification
Smoking risk analysis
Security Features
JWT Authentication
Role-based access
Encrypted medical records
HIPAA-style privacy protection
Architecture Idea
Frontend (Vue/React)
↓
REST API Backend
↓
AI Microservice
↓
Database + Cloud Storage
Best Tech Stack Combination
Option 1 (Recommended)
Vue.js
Node.js Express
Python Flask AI Service
MySQL
TensorFlow
Option 2
Laravel
Python AI API
PostgreSQL
React
Future Improvements
AI voice diagnosis assistant
Blockchain medical records
Integration with hospital systems
Real-time telemedicine
Good Final Year Project Title

“Development of an AI-Powered Lung Cancer Diagnostic and Prediction System Using Deep Learning”

Or

“Deep Learning-Based Lung Cancer Detection System Using CT Scan Images”

Skills: Java, Python, Machine Learning (ML), MySQL, PostgreSQL, Image Processing, OpenCV, Deep Learning

Cancer Detection Using Machine Learning

## Project Overview

I have developed a Machine Learning based Lung Cancer Detection system that predicts the likelihood of lung cancer using patient medical parameters and data analysis techniques.

This project is suitable for:

* Final year engineering students
* Academic submissions
* Research work
* Machine Learning learning purposes
* Healthcare AI projects

## Features

* Data preprocessing and cleaning
* Exploratory Data Analysis (EDA)
* Machine Learning model training and testing
* Accuracy evaluation and prediction system
* Visualization graphs and reports
* Easy-to-understand code structure
* User-friendly implementation

## Technologies Used

* Python
* Scikit-learn
* Pandas
* NumPy
* Matplotlib
* Jupyter Notebook

## Deliverables

* Complete source code
* Trained ML model
* Dataset (if shareable)
* Documentation/report
* PPT for presentation
* Installation and execution guide

## Why Choose This Project?

* Clean and well-commented code
* Beginner-friendly structure
* Ready for academic use
* Customization support available
* Fast delivery

## Extras Available

* Web application deployment
* GUI integration
* Deep Learning enhancement
* Accuracy improvement
* Project explanation support



## Suggested Pricing

* Basic Project: $30
* With Documentation & PPT: $50
* Full Customization & Support: $80+

## Delivery Time

1–3 Days

**Project Title Suggestions**

* Lung Cancer Detection Using Machine Learning in Python
* Healthcare ML Project with Source Code
* AI-Based Lung Cancer Prediction System
* Final Year Machine Learning Medical Project

Skills: Python, Powerpoint, Machine Learning (ML), Data Mining, Hadoop, Artificial Intelligence, Visualization, NumPy, Data Analysis, Pandas

We are looking for a hands-on Senior Computer Vision Architect to scale our live multi-camera video analytics platform from 100 to 1,000+ cameras across NVIDIA DeepStream and distributed edge device architectures.

The ideal candidate must have strong production experience with NVIDIA DeepStream SDK, TensorRT optimization, edge AI deployments, and large-scale camera infrastructure.

Responsibilities:
• Design and manage a scalable 1,000-camera architecture with high availability and low latency
• Optimize DeepStream pipelines (NvInfer, NvTracker, NVDEC, GStreamer) for stable FPS under heavy load
• Deploy an manage CV models on Jetson or equivalent edge devices
• Fine-tune YOLO models and export optimized TensorRT engines
• Build and improve infrastructure using Docker, Kafka/MQTT, GPU clusters, load balancing, and autoscaling
• Improve real-time analytics dashboards and monitoring systems
• Document deployment workflows and architecture decisions

Required Skills:
• NVIDIA DeepStream SDK
• YOLO v5/v8/v9
• TensorRT, ONNX, TAO Toolkit
• Jetson Nano/Orin
• Docker, Kafka/MQTT, Nginx
• RTSP, ONVIF, H.264/H.265
• Python and C++

Experience Required: 5+ years in computer vision engineering
• 3+ years hands-on DeepStream experience
• Experience handling 250+ camera deployments
• Strong TensorRT optimization experience

To Apply:
Please include GitHub repositories, portfolio links, or case studies showing DeepStream pipelines or large-scale camera deployments.

Skills: Python, Machine Learning (ML), Artificial Intelligence, Docker, OpenCV, Computer Vision, Deep Learning, YOLO

# Lung Cancer Detection Using Machine Learning

## Project Overview

I have developed a Machine Learning based Lung Cancer Detection system that predicts the likelihood of lung cancer using patient medical parameters and data analysis techniques.

This project is suitable for:

* Final year engineering students
* Academic submissions
* Research work
* Machine Learning learning purposes
* Healthcare AI projects

## Features

* Data preprocessing and cleaning
* Exploratory Data Analysis (EDA)
* Machine Learning model training and testing
* Accuracy evaluation and prediction system
* Visualization graphs and reports
* Easy-to-understand code structure
* User-friendly implementation

## Technologies Used

* Python
* Scikit-learn
* Pandas
* NumPy
* Matplotlib
* Jupyter Notebook

## Deliverables

* Complete source code
* Trained ML model
* Dataset (if shareable)
* Documentation/report
* PPT for presentation
* Installation and execution guide

## Why Choose This Project?

* Clean and well-commented code
* Beginner-friendly structure
* Ready for academic use
* Customization support available
* Fast delivery

## Extras Available

* Web application deployment
* GUI integration
* Deep Learning enhancement
* Accuracy improvement
* Project explanation support



## Suggested Pricing

* Basic Project: $30
* With Documentation & PPT: $50
* Full Customization & Support: $80+

## Delivery Time

1–3 Days

**Project Title Suggestions**

* Lung Cancer Detection Using Machine Learning in Python
* Healthcare ML Project with Source Code
* AI-Based Lung Cancer Prediction System
* Final Year Machine Learning Medical Project

Skills: Python, Matlab and Mathematica, Machine Learning (ML), Hadoop, NumPy, Data Visualization, Data Analysis, Pandas

I’m looking for an engineer to take a close look at my current Bedrock agent core gateway, spot the pain-points, and deliver a clean set of improvements. The codebase is stable but i need few changes Changes include 1. Create a RCA ( Root cause analysis agent ) 2. Build and deploy into Agent Runtime 3. Use Observability for monitoring Otel metrics logs and tracing 4. Use Memory for caching 5. Add skills to the Agent RCA( Root cause analysis) agent: currently RCA is triggered based on failure signals, not regression against a baseline. The regression-based RCA (k6 + SpeedCurve compares against baseline if diff found then mapping to git diffs to see what was changed , go through the commits , this is not fully implemented yet. The RCA also looks into datadog metrics and uses AI to detect anomalies . I want above to be implemented

Hello Upwork Freelancers,
Looking for an up and coming talented AI programmer to help introduce a new AI powered webportal service for use by contractors and government agency staff to help prepare Traffic Control Plans (TCPs) online. TCPs are normally prepared by engineers or trained technicians in Compated Aided Drafting (CAD) software and are usually prepared as site-specific plans using a satellite image background.

The specific service we're looking to introduce would allow a user to draw over a satellite image the location of their work zone in the roadway. Following the work zone location a to-be-build AI StreetScan tool would create a basemap of the roadway that includes lane lines, curb lines, medians, train tracks, and pavement messages (arrows, STOP, YIELD, etc.). After the basemap is drawn the User would be asked to verify the basemap and confirm the roadway speed limit. The AI-TCP tool would then draw the traffic control devices on the basemap including traffic cones, arrow boards, construction zone signs, etc. A tool to edit the drawing elements created must be provided. The final TCPs to the Users should be provided as a PDF document depending on page size selected and by exportable as a CAD file in DXF, DWG, or DGN format if possible. The preferred scale for TCP drawings would be either 1" = 40' (preferred) or 1" = 50'.

We have thousands of TCPs drawn to help in training the software.

See attached project work scope and description.

Auto reply messages that clearly show the work scope was not reviewed will not be considered nor responses provided. We seek to bring on a team within two weeks to get this project started.

Provide a cost estimate and schedule for development. We seek the platform to be built on Digital Ocean and will provide a droplet server for use. All software code must be uploaded to a provided GitHub account and all software developed will become property of Traffic Patterns LLC - Danville, CA.

Experienced AI/computer-vision developer or engineering team required for an intelligent transport infrastructure project focused on real-time roadside monitoring and vehicle/sensor analysis. Tasks include tyre tread assessment, low-speed vehicle detection, sensor integration, ANPR, embedded systems, machine-learning model development, and dashboarding for data analysis and visualization. Familiarity with OpenCV, deep learning frameworks, edge deployment, and smart-transport technologies is essential. A detailed brief and technical specification will be provided upon engagement and NDA. Prior relevant portfolio required.

Build according to www.aitoolsengine.app (visualization) and as discussed. See attached build plan and timeline.

The goal is to build a hardware-software solution that sits in a consumer’s living room and measures "eyes-on-glass" attention in real-time.
We are looking for an expert in NVIDIA Jetson development to build the core Computer Vision and Audio ACR (Automatic Content Recognition) pipeline. This device will process all data locally at the edge to ensure privacy, sending only anonymized attention metadata to our cloud bidding engine.

Key Responsibilities
Edge AI Pipeline: Develop a real-time multimodal pipeline using the NVIDIA DeepStream SDK on Jetson Orin Nano/NX.
Computer Vision Implementation: Integrate and fine-tune models for:
Person Detection & Counting: (Co-viewing metrics).
Gaze Estimation: Measuring if a viewer's eyes are directed at the "TV zone" vs. secondary devices.
Pose Estimation: Distinguishing between active viewers and people passing through the room.
Model Optimization: Use the NVIDIA TAO Toolkit to fine-tune pre-trained models for low-light (NIR) and long-range living room environments.
Model Compression: Perform pruning and quantization (INT8/FP16) to ensure the pipeline runs efficiently under 15W.
Audio ACR Integration: Implement passive audio fingerprinting (e.g., Chromaprint or neural hashing) to identify the content playing on the TV.
Privacy & Deployment: Ensure the system operates with "Zero-Video Storage"—processing frames in RAM and deleting them immediately after inference.

We are seeking a skilled and forward-thinking AI Developer to build a custom AI system for an innovative project. This role involves designing, developing, and implementing intelligent solutions tailored to specific business needs. If you are passionate about artificial intelligence and enjoy building impactful systems from scratch, we would love to hear from you.

Key Responsibilities:
Design and develop a custom AI-powered system
Build, train, and optimize machine learning models
Integrate AI solutions with existing platforms and APIs
Work closely with stakeholders to understand project requirements
Ensure scalability, performance, and security of the system
Provide documentation and technical support

We are looking for an experienced computer vision engineer for an automated grocery store checkout project.

(Yes, Amazon and Aldi tried this, and we think we can crack it nevertheless)

I am looking for an experienced Quantitative Developer / Systems Architect to design and deploy a robust, low-latency algorithmic trading platform. The goal is to move from manual execution to a fully automated system that handles data ingestion, strategy execution, and risk management.

Scope of Work
Infrastructure Setup: Configure a secure, Homelab / cloud-based environment (Oracle) optimized for 24/7 uptime.

Exchange Integration: Connect via REST/WebSocket APIs to CMC Market, Interactive Brokers, Td waterhouse, Questrade.

Strategy Implementation: Translate my existing logic into clean, backtested code. The strategy involves Mean Reversion, Momentum, Arbitrage or others .

Risk Management Engine: Develop "Circuit Breakers" including Hard Stop Losses, Max Daily Drawdown limits, and position sizing logic.

Dashboard/Logging: Set up a basic UI or Telegram/Discord bot for real-time trade alerts and performance tracking.

Technical Requirements
Languages: Proficiency in Python (Pandas, NumPy, TA-Lib) or C++.

Frameworks: Experience with Backtrader, Lean (QuantConnect), or CCXT.

DevOps: Docker, Linux server management, and basic cybersecurity protocols.

Database: Experience with Time-Series Databases (InfluxDB, Kdb+) is a plus.

Preferred Qualifications
Proven track record of building "production-ready" bots (not just backtesting scripts).

Deep understanding of slippage, latency, and order book dynamics.

Ability to explain complex logic simply and provide clean, documented code.

Application Questions
Which exchanges/brokers have you successfully integrated with using Python?

How do you handle API disconnections or "flash crash" scenarios in your code?

Can you provide a brief overview of a previous trading system you built and the tech stack used?

What is your preferred method for backtesting to ensure "Look-Ahead Bias" is avoided?

We are looking for a specialist, team, or company that can provide or create a drone detection dataset and help us build a reliable AI model for Daylight and thermal camera video.
The main requirement is high-quality data.

We are ready to buy a suitable drone detection dataset if it includes real-world drone footage and can legally be used for model training and commercial/internal deployment.
We are also open to custom synthetic dataset creation. Professionals who can create realistic drone detection datasets using Unreal Engine, AirSim or similar tools are welcome.

The dataset should ideally include:
Daytime drone footage
Night / thermal / IR drone footage
Small drones at long distance
Real or realistic synthetic outdoor backgrounds
Birds, aircraft, insects, clouds, trees, buildings, and other negative examples
Proper object detection annotations
We are not looking only for someone who can download public datasets and train YOLO. Public datasets may be used, but we need something more advanced: private data, commercial data, synthetic data, custom collection, annotation, or a full dataset/model development workflow.

The final model should be suitable for our technical environment. Camera resolution, lens, DRI requirements, detection distance, hardware platform, and export format will be discussed during the technical conversation.

Please apply if you can clearly explain:
What drone detection dataset you can provide or create
Whether it is public, private, proprietary, commercial, or synthetic
Whether it includes thermal / IR data
Dataset size and annotation format
Your synthetic data generation approach, if relevant
Whether you can support ONNX / TensorRT / Jetson / DeepStream deployment (optional)

• Have you previously worked on drone detection area?
• Do you have access to drone detection datasets (day or thermal) that are not only public/open-source?
• How do you handle very small drones in the image?
• What metric would you use to provide the model is reliable?
• Can you provide a demo on sample videos?

Request for Proposal: Enterprise Executive Agentic AI OS
1. System Overview
We are looking to build a secure, hybrid (local/cloud) Agentic AI Operating System designed to act as a comprehensive "co-pilot" for a Director-level executive at a major enterprise AI infrastructure company. The executive oversees product platform strategy for the Retail and CPG sectors. Because of the high volume of technical coordination, ecosystem management, and strategic planning required, this system must autonomously draft communications, synthesize technical architectures, generate strategic roadmaps, and orchestrate cross-functional workflows, subject to a "human-in-the-loop" approval process.

2. Core Modules & Capabilities

Module A: The Communications & Coordination Engine


Email & Chat Autonomous Drafting: The agent must integrate with enterprise communication tools (Outlook/Gmail, Slack/Teams). It should read incoming threads, extract the core intent, and automatically draft professional, context-aware replies.


Tone Matching: The system must analyze past communications to adopt a definitive, authoritative, yet collaborative managerial tone.


Human-Like Nuance & Tone Masking: The agent must be explicitly engineered to avoid standard, recognizable LLM phrasing, vocabulary, and formatting (e.g., avoiding words like "delve," "crucial," "tapestry," or overly structured bullet points when a quick sentence will do).


Dynamic Persona Emulation: The system must be able to mimic natural human communication quirks. It should allow the user to adjust the "formality slider" (e.g., using colloquialisms in Slack, but maintaining formal executive presence in email) and be capable of learning the user's specific stylistic habits (e.g., preferred sign-offs, sentence lengths, and common industry abbreviations).


Contextual Brevity: High-level executives do not write long, overly explanatory emails. The agent must default to extreme brevity and directness, mimicking the fast-paced, high-efficiency communication style of a busy technical leader.


Calendar & Meeting Orchestration: The agent must automatically prioritize meeting requests based on stakeholder importance, suggest agendas based on previous communications, and generate pre-read summaries for the executive 10 minutes before every meeting.

Module B: Strategic Planning & Goal Setting


OKR & Strategy Generation: The system must be able to ingest raw meeting transcripts, partner feedback, and market research, and output structured Objectives and Key Results (OKRs).


Roadmap Synthesis: When provided with disparate technical updates from engineering teams, the agent must generate cohesive, executive-level product roadmaps and slide deck outlines tailored for external partners (ISVs, GSIs) and internal leadership.


Business Value Translation: The agent must be trained to automatically translate highly technical feature releases into "business value" pitches for retail and CPG clients (e.g., focusing on latency, throughput, and cost efficiency).

Module C: Technical Architecture & "Proxy Engineering"


(Note to Developers: The user of this OS needs the system to heavily bridge the gap between business and engineering.)


Reference Architecture Generation: The system must utilize RAG (Retrieval-Augmented Generation) connected to an internal knowledge base of enterprise AI capabilities (e.g., GPU-accelerated computing, LLM inference, data pipelines). When asked, it must generate technical reference architectures, system design documents, and API integration guides.


Code & Script Review: The agent should be able to review technical documentation or code snippets provided by engineering teams and summarize the "impact on the customer experience" in plain English.

Module D: Ecosystem & Partner Management


Partner Intelligence: The system must track relationships with Global Systems Integrators (GSIs) and Independent Software Vendors (ISVs).


Engagement Tracking: It should prompt the executive when it is time to follow up with a partner and automatically draft the follow-up communication based on the last known blocker or milestone.

3. Technical & Architectural Requirements


Hybrid Deployment (Local + Cloud): The system should utilize a lightweight local agent (for secure, low-latency access to local file systems and immediate context) and securely offload heavy compute/LLM processing to a cloud environment.


Agentic Orchestration: The system must use an orchestration framework (e.g., LangGraph, AutoGen) to manage multiple specialized sub-agents (e.g., an Email Agent, a Strategy Agent, a Technical Architect Agent) that can pass context back and forth.


RAG (Retrieval-Augmented Generation): The system must include a secure vector database to index the executive's local files, past emails, company product documentation, and industry (Retail/CPG) trends.

4. Security & Guardrails (Crucial)


Human-in-the-Loop (HITL) Mandate: The agent must not send any email, schedule any meeting, or modify any file without explicit, one-click approval from the user.


Hallucination Mitigation: The system must utilize input and output guardrails to prevent it from inventing technical capabilities or making false commitments to partners. It must provide citations to internal documents for any technical claim it makes in a drafted email or strategy doc.


State Rollback: The system must maintain an audit log of every action taken and allow for immediate rollback of any local file modifications.

5. Project Logistics & Requirements


Data Privacy & LLM Hosting: Because this AI will read highly confidential enterprise emails and strategic documents, the developer cannot use standard consumer API endpoints that train on user data. You must utilize enterprise-grade APIs (e.g., OpenAI Enterprise, Anthropic's Console with zero data retention) or fully local open-source models (e.g., Llama-3).


Phased Rollout: We require developers to structure their proposals around a phased build:


Phase 1: Email & Chat Communications Engine (Module A).


Phase 2: RAG Database & Proxy Engineering (Module C).


Phase 3: Strategic Planning & Ecosystem Tracking (Modules B & D).

Proposal Submission: Please provide your estimated timeline and cost structure broken down by the phases listed above.


Maintenance & Handoff: At the conclusion of the build, the code must be fully packaged with clear documentation so that I can run the system independently on my machine. Additionally, I require an ongoing support/maintenance arrangement so I can reach out to you directly to fix the system if any API endpoints update or break.

We need an experienced AI agent developer to build a custom autonomous agent that can reason, use tools, and complete multi-step tasks on our behalf. This is a defined, fixed-price project with clear deliverables — we want someone who has shipped production AI agents before, not someone learning on our budget.

━━ PROJECT SCOPE ━━

→ Design and build a multi-step AI agent using LangChain, LangGraph, or CrewAI
→ Agent must use tool-calling to interact with external APIs and our internal database
→ Implement RAG (Retrieval-Augmented Generation) using our document corpus (PDF + web sources)
→ Vector store setup using Pinecone or Qdrant for semantic search
→ Memory management — short-term conversation memory + long-term persistent context
→ Deploy the agent via n8n or a lightweight API endpoint (FastAPI / Node.js)
→ Full documentation + handover call

━━ STACK WE PREFER ━━

OpenAI GPT-4o · Anthropic Claude · LangChain / LangGraph · Pinecone or Qdrant · n8n or Make · Python · FastAPI · Supabase or PostgreSQL

━━ YOU MUST HAVE ━━

✔ Proven experience building production AI agents (not just chatbots — actual tool-calling agents)
✔ Deep knowledge of prompt engineering, RAG architecture, and vector embeddings
✔ Experience with LangChain, LangGraph, CrewAI, or equivalent agent frameworks
✔ Can explain your architecture decisions in plain English
✔ Portfolio or GitHub showing real AI agent work

We are seeking an AI Solutions Architect to lead the development of our Agentic AI platform. The role involves designing and implementing AI solutions, collaborating with cross-functional teams, and ensuring seamless integration with our existing systems. The ideal candidate will have experience in Full Stack Python and React, and be able to manage both front-end and back-end development tasks.

# AI Automation Consultant – Claude.ai Agent Development

**Job Title:**
AI Automation Consultant / Claude.ai Agent Developer

**Job Type:**
Independent Contractor / Freelance

**Project Overview:**
We are seeking an experienced AI automation consultant who specializes in building intelligent workflow agents using [Claude.ai](https://claude.ai?utm_source=chatgpt.com) and related AI tools. Our company manages high volumes of pricing, sourcing, and operational tasks across multiple vendor and backend systems, and we are looking to streamline these workflows using AI-powered agents and automation.

The ideal candidate should understand how to build practical AI agents that can interact with multiple websites, gather information, compare data, and assist our internal team with repetitive operational tasks.

## What We Need

Currently, our staff manually logs into multiple vendor and supplier websites to:

* Check product availability
* Compare pricing
* Verify stock levels
* Gather order information
* Assist with quoting and procurement workflows

In many cases, our team must access 5 or more different platforms for a single order. We want to automate and streamline these processes using AI agents and backend automation systems.

## Scope of Work

We are looking for someone who can:

* Build AI agents using Claude.ai and/or similar AI frameworks
* Create automations that interact with multiple websites and backend systems
* Help streamline procurement and quoting workflows
* Build tools that can:

* Log into vendor portals
* Pull inventory and pricing data
* Compare supplier information
* Present consolidated results to our team
* Recommend the best architecture and AI stack for long-term scalability
* Help integrate APIs, browser automation, and AI workflows where needed
* Potentially integrate with internal systems such as spreadsheets, portals, or databases

## Preferred Skills

* Experience with Claude.ai, AI agents, and workflow automation
* Strong understanding of browser automation tools
* Experience with:

* Python
* APIs
* Web scraping/automation
* AI orchestration tools
* Backend integrations
* Familiarity with tools such as:

* Playwright
* Selenium
* LangChain
* MCP servers
* Zapier
* Make.com
* OpenAI / Anthropic APIs
* Experience building AI-powered business process automations
* Strong communication and consulting skills

## Ideal Candidate

We are looking for someone who is:

* Practical and solutions-oriented
* Experienced in operational workflow automation
* Able to recommend scalable AI solutions
* Comfortable working with multiple systems and vendor platforms
* Able to think like both an engineer and business consultant

## To Apply

Please provide:

* Examples of AI agents or automation systems you have built
* Relevant experience with Claude.ai or AI workflow automation
* Your recommended approach for this type of project
* Estimated timeline and hourly/project rates
* Any case studies or portfolio examples