AI/ML Engineer specializing in Agentic AI Systems, Deep Learning, and Large Language Models. 4+ years of experience delivering 100+ AI/ML projects with 85–96% accuracy in real-world applications.
I'm Azhar Ali, a Computer Science graduate from Sukkur IBA University (GPA: 3.5/4.0) with a deep passion for Artificial Intelligence. I specialize in building autonomous AI agents using LangChain/LangGraph, implementing RAG systems with vector databases, and developing production-grade ML models for Computer Vision and NLP.
With 4+ years of freelance experience on platforms like Kwork, Fiverr, and Upwork, I've delivered 100+ AI/ML projects achieving a 98% client satisfaction rate. I've architected multi-agent systems, built RAG pipelines processing 100K+ documents, and deployed edge AI models achieving 30 FPS on embedded devices.
Get In TouchTechnologies and tools I work with
LangChain, LangGraph, AutoGPT, Multi-Agent Systems, Agent Orchestration, Tool Use, OpenAI API, Claude API, Hugging Face Transformers
PyTorch, TensorFlow, Keras, OpenCV, YOLO, ResNet, EfficientNet, Object Detection, Image Classification, Facial Recognition
FAISS, ChromaDB, Pinecone, Weaviate, Semantic Search, Embeddings, Vector Indexing, Document Retrieval Pipelines
BERT, GPT, Transformers, Sentiment Analysis, Named Entity Recognition, Text Generation, Tokenization, Fine-tuning
AWS SageMaker, Google Cloud AI, Docker, Kubernetes, FastAPI, Flask, MLflow, Weights & Biases, TensorBoard
React, Node.js, Express.js, Next.js, Spring Boot, MongoDB, PostgreSQL, REST APIs, HTML/CSS/JS, Flutter
Professional journey & achievements
4+ years, 50+ projects, consistent 5-star ratings on Kwork
Machine Learning, Deep Learning & Advanced CV Certifications
Published content on Agentic AI & LangGraph reaching 1000+ developers
32 public repositories on GitHub — CAL.com contributor
Showcasing AI/ML solutions that solve real problems
Multi-agent system with LangGraph for autonomous research achieving 85% accuracy. Semantic search with FAISS on 100K+ docs with sub-second response. Agent workflows with state management and self-reflection.
Conversational AI with multi-turn context using Claude Sonnet 4. Fallback system with 99.9% uptime serving 500+ monthly visitors. Built with LangChain for context-aware responses.
Fine-tuned BERT achieving 94% accuracy. Scalable pipeline processing 1K+ posts daily with <200ms latency. ONNX optimization for production. Real-time dashboard for monitoring.
CNN with facial landmarks achieving 96% accuracy. EAR/MAR features for reliable detection. Edge deployment on Raspberry Pi with quantization achieving 30 FPS and <100ms latency.
ResNet-50 / EfficientNet achieving 94% accuracy on 5K images. Data augmentation, mixed precision training. Deployed with TensorFlow Serving REST API for production inference.
Hybrid recommendation engine combining Item-Based Collaborative Filtering and SVD Matrix Factorization. Cold-start handling, explainability module, and Gradio web interface.
Cross-platform staff management system built with Flutter & Node.js. Features QR-based clock-in/out, real-time attendance tracking, rota scheduling, and role-based admin/staff dashboards. Deployed on Netlify & Render.
Enterprise Customer Relationship Management system built with Spring Boot backend and React frontend. Features customer data management, interaction tracking, analytics dashboards, and scalable REST API architecture.
Let's work together to build something amazing
I'm always open to discussing new projects, creative ideas, or opportunities to be part of your vision. Whether it's building an AI agent, training an ML model, or developing a full-stack application — let's connect!