engineer:
name: "Haseeb Ul Hassan"
role: "AI / Machine Learning Engineer"
experience: "2+ years"
location: "Riyadh, Saudi Arabia"
focus:
- Production Computer Vision systems (detection, OCR, OBB)
- NLP & structured information extraction
- Generative AI / LLM-powered pipelines (RAG, agents)
- End-to-end MLOps: from notebook to deployed service
philosophy: >
I engineer AI systems the way a software engineer ships products -
reliable, observable, and built to survive contact with real data.I'm an AI/ML Engineer with 2+ years of experience designing and deploying production AI systems across Computer Vision, NLP, and Generative AI. I built ForecastIQ, an end-to-end demand-forecasting pipeline that replaced manual spreadsheet workflows and now saves $200K+ annually, and I've delivered custom CV / OCR / LLM solutions for enterprise clients - from rotated-object detection models to LLM-refined data-extraction pipelines.
I care about the full lifecycle of an ML system - not just training a model, but shipping it behind an API, monitoring it, and keeping it correct in production. I approach AI engineering with a software-engineering mindset: version-controlled experiments, containerized services, and pipelines built to be maintained, not just demoed.
🎯 Open To: AI/ML Engineering roles · Computer Vision & NLP projects · Applied GenAI / RAG systems · Freelance & contract ML engineering · Co-founding early-stage AI products
| Domain | Proficiency | Details |
|---|---|---|
| Computer Vision | ⭐⭐⭐⭐⭐ | Custom-trained YOLO / YOLO-OBB detection models, data augmentation, hyperparameter tuning, segmentation vs. detection benchmarking |
| OCR & Document AI | ⭐⭐⭐⭐⭐ | Tesseract, EasyOCR, PaddleOCR, Google Cloud Vision, LayoutLMv3 fine-tuning for document sequence classification |
| NLP | ⭐⭐⭐⭐☆ | Custom NER models, structured field classification, LLM-based post-processing pipelines |
| Generative AI / LLMs | ⭐⭐⭐⭐⭐ | RAG pipelines (ChromaDB, Pinecone), OpenAI & Gemini SDKs, agentic workflows, prompt engineering |
| Forecasting & Planning | ⭐⭐⭐⭐⭐ | SKU-level demand forecasting, rule-driven purchasing logic, automated scheduled pipelines |
| MLOps & Deployment | ⭐⭐⭐⭐☆ | FastAPI services, Docker containers, VPN-secured VM deployment, W&B / MLflow experiment tracking |
📦 ForecastIQ — Demand Planning & Forecasting Pipeline
Built a flexible-horizon (24–54 week) SKU-level forecasting pipeline with a Python planning engine that applies business rules extracted from structured notes files to derive purchasing formulas for the client's procurement team. Deployed on a client-provisioned VM accessed over VPN, served via FastAPI with automated daily execution.
| Aspect | Details |
|---|---|
| Stack | Python, FastAPI, Pandas, NumPy, PostgreSQL, Docker |
| Scale | SKU-level forecasts across a 24–54 week flexible horizon |
| Performance | Automated daily execution with scheduled pipeline runs |
| Security | VPN-restricted access on a client-provisioned VM |
| Impact | Replaced manual spreadsheet workflows — $200K+ saved annually in labor and reduced excess inventory |
| Repository | Private — Enterprise Client Deliverable |
🏷️ Enterprise Image-Based Cataloging & Data Extraction System
Delivered an automated image-based cataloging and data-extraction system for a major US enterprise client, combining a custom-trained YOLO-OBB model, OCR, and LLM-based classification to reduce manual entry and improve data accuracy.
| Aspect | Details |
|---|---|
| Stack | YOLO-OBB, OCR, LLM Classification, Python |
| Scale | Enterprise-level cataloging workflow for a major US client |
| Performance | Reduced manual entry time via automated extraction |
| Security | Client-managed enterprise deployment |
| Impact | Improved cataloging data accuracy and reduced manual workload |
| Repository | Private — Enterprise Client Deliverable |
AI / ML Engineer @ VisionX
Aug 2024 — Aug 2026 · New York, USA (Remote)
Leading ML initiatives across Forecasting, Automation, Computer Vision, and Generative AI deployment workstreams for enterprise clients.
| Recognition | Details |
|---|---|
| 🥇 Presidential Gold Medalist | Awarded for top academic standing — CGPA 3.77/4.0, Institute of Space Technology |
| 💰 $200K+ Annual Savings | Delivered via the ForecastIQ demand-forecasting pipeline at VisionX |
| 🎓 SCE Registered | Computer Science Specialist, Saudi Council of Engineers |
current_focus:
learning:
- Advanced agentic LLM architectures & multi-agent orchestration
- Model optimization & efficient inference (vLLM, quantization)
building:
- Production-grade RAG and agentic pipelines for enterprise clients
- Scalable CV + OCR + LLM cataloging systems
exploring:
- Fine-tuning open-weight LLMs for domain-specific tasks
- Edge deployment of vision models
open_to:
- AI/ML Engineering roles
- Applied GenAI & RAG consulting
- Early-stage AI product collaboration
"Good models are trained. Great systems are engineered."