I'm an AI & LLM Engineer with 5+ years of experience designing and shipping production-grade AI systems — not demos. My work spans RAG pipelines, multi-agent workflows, LLM fine-tuning, computer vision, and MLOps infrastructure. I care deeply about measurable outcomes: retrieval hit@10, p95 latency, cost-per-request. If I can't measure it, I'm not done.

Currently building agentic AI systems at Turing and contributing to A.Team missions in AI & data science.

• RAG Systems — LangChain + FAISS + Neo4j Vector, deployed on GCP Cloud Run. Turning 3-4 hours of manual document analysis into 2-minute AI-powered answers.
• Multi-Agent Pipelines — Generator-Validator architectures, Plan-Execute frameworks, Think-Act-Observe loops. Agents that are reliable, not just impressive.
• LLM Fine-Tuning — SFT models for code translation and domain reasoning. Taxonomy-aligned training datasets for classification, generation, and reasoning tasks.
• Agentic Evaluation Frameworks — Designed and evaluated 850+ long-horizon trajectories (100–200+ actions each) at Turing. Built QA rubrics that classify outputs into Critical Mistakes, Bad Side Effects, and Misunderstandings.
• Computer Vision — YOLOv8/v5 object detection, OCR pipelines (Tesseract), NER systems achieving 95% OCR accuracy and 90% NER precision.

AI & LLM

Backend & APIs

Databases & Vector Stores

Cloud & DevOps

Frontend

AI Tools I Use Daily

• Remote AI/LLM engineering roles
• Multi-agent and RAG system projects
• LLM evaluation and fine-tuning collaborations
• Computer Vision + NLP hybrid projects

Based in Pakistan 🇵🇰 | Available globally | Fluent in English