AI Engineer and researcher focused on practical, safety-aware AI systems.
I build end-to-end AI products across healthcare, computer vision, retrieval systems, and LLM infrastructure. My work usually sits at the point where models have to survive real product constraints: privacy, evaluation, cost, review workflows, deployment, and maintainability.
- Production LLM systems: RAG, routing, evals, memory, safety layers, and model cost control.
- Healthcare AI: medical imaging, clinical safety patterns, explainability, and auditability.
- Applied research: translating prototypes into workflows that can be tested, reviewed, and improved.
- Developer tooling: small starter kits that make serious AI engineering patterns easier to study.
| Project | What it shows |
|---|---|
| thesis | M.S. thesis workspace for CYCLES-based crop-management reinforcement learning, reporting, and dashboard artifacts. |
| not-loom | Local-first screen, mic, and webcam recording with a React/Vite frontend and optional Flask native recorder. |
| LifeMap | A React app that maps personal trajectory using W6H questioning and Hype Cycle-style analysis. |
| production-rag-toolkit | RAG building blocks: extraction, hybrid retrieval, query expansion, reranking, and eval gates. |
| llm-gateway-starter | FastAPI gateway pattern for routing, rate limits, caching, fallback, and cost caps. |
| healthcare-ai-safety-patterns | Clinical AI safety pipeline with risk classification, output validation, and redacted audit logging. |
| medical-image-ai-starter | Medical imaging starter kit with preprocessing, Grad-CAM, EfficientNet fine-tuning, and U-Net utilities. |
| fintech-ai-compliance-kit | Compliance-aware AI patterns for model cards, audit trails, explanations, fraud, and KYC demos. |
| token-economics-calculator | Python and spreadsheet tools for estimating LLM feature cost, margin, caching, and price-shock scenarios. |
- M.S. Artificial Intelligence, NUST.
- AI engineering experience across healthcare, NLP, computer vision, and product systems.
- Published research in medical AI, including osteoarthritis grading with deep learning.
- Kaggle Expert with experience building datasets, notebooks, and applied ML workflows.
Python, PyTorch, TensorFlow, FastAPI, React, TypeScript, Vite, Docker, PostgreSQL, MongoDB, AWS, Azure, GCP, OpenAI APIs, retrieval systems, evaluation workflows, and medical image processing.
- LinkedIn: haidermasood99
- Email: haidermasood310@gmail.com
I keep this profile focused on projects that demonstrate shipped engineering judgment rather than just experiments. Older coursework, forks, and raw data repos are being cleaned up or archived.