Engineer first. Researcher on the side.
I build agents and retrieval systems that companies rely on, and I explain new AI research on YouTube. I am looking for a full-time AI Engineer role and I am ready to relocate.
Huzaifa Tahir at a glance.
- Role
- Lead AI Engineer, SyntaxBrain
- Based in
- Lahore, Pakistan · PKT (UTC+5)
- Experience
- 3+ years in production AI, 5 companies
- Specialises in
- Agentic AI, LangGraph, CrewAI, MCP, RAG, LLM fine-tuning
- Notable work
- RAG over 10M+ legal documents · 90% payroll time cut
- Education
- BSc Computer Science, Superior University Lahore
- Open to
- Full-time roles in DE · NL · UK · UAE, 2-week notice
Meet Huzaifa
A two-minute introduction: who I am, what I build and how I work with teams.
A full-time AI Engineer role, with relocation.
Huzaifa Tahir is an AI Engineer from Lahore, Pakistan, with 3+ years building production agentic AI: LangGraph and CrewAI multi-agent systems, MCP servers and RAG pipelines. He is Lead AI Engineer at SyntaxBrain, previously at Musketeers Tech, Fabtechsol and Mexa Solutions, authored "Mastering Ensemble Techniques", and explains AI research papers on YouTube.
Experience
Five companies, all shipping AI to real users. Currently Lead AI Engineer at SyntaxBrain.
Lead AI Engineer
SyntaxBrain · Lahore- Leads a legal AI platform for a UK law firm with 15+ lawyers as lead engineer and project manager: retrieval over 10M+ documents plus fine-tuning of open-source Llama 3.1 models.
- Owns architecture and delivery for agentic and RAG systems, from scoping with clients to production.
AI Engineer
Musketeers Tech · Lahore- Built a payroll and salary-slip automation system that cut monthly processing time by 90% for 800+ employees.
- Architected a Deep Research Sales Intelligence Agent with CrewAI and FastAPI that researches and qualifies B2B leads end to end; top-ranked among client deliverables.
- Developed custom MCP servers integrated with OpenAI, reducing agent integration overhead across projects.
AI Engineer
Fabtechsol · Lahore- Led end-to-end integration of the B-Master multi-agent platform with CrewAI, LangChain and LangGraph.
- Fine-tuned GPT-4 Vision for car make and year detection at 74% classification accuracy, feeding automated price estimation.
- Built tax-strategy agents with the LangChain SQL Agent, OpenAI and Tavily Search to cut analyst research time.
Generative AI Engineer
Mexa Solutions · Lahore- Built an Interview Copilot with OpenAI, Deepgram real-time speech-to-text and Groq, with sub-second response latency.
- Implemented RAG pipelines with LangChain and LlamaIndex for cybersecurity document analysis and research tools.
- Shipped a finance chatbot on Streamlit and Snowflake for natural-language queries over the data warehouse.
Associate Python Developer
FalconXoft · Remote- Built booking logic and backend APIs for a flight reservation system on third-party flight APIs.
- Developed a Django app for real-time face-emotion recognition with CNNs.
- Wrote a Best Seller Rank scraper that informed e-commerce merchandising decisions.
BSc Computer Science
Superior University Lahore, 2019 to 2023
Machine Learning · NLP Specializations
Coursera, DeepLearning.AI, 2023
Mastering Ensemble Techniques
The Art and Science of Smarter Machine Learning: bagging, boosting and stacking in practice.
Research and teaching
I read new AI papers so builders don't have to, and write long-form guides from production work.
SyntaxBrain
Frequently asked
Who is Huzaifa Tahir?
Huzaifa Tahir is an AI Engineer from Lahore, Pakistan, with 3+ years building production agentic AI: LangGraph and CrewAI multi-agent systems, MCP servers and RAG pipelines. He is Lead AI Engineer at SyntaxBrain and previously worked at Musketeers Tech, Fabtechsol and Mexa Solutions.
Is Huzaifa Tahir open to new roles?
Yes. He is open to full-time AI Engineer, Applied AI Engineer, LLM Engineer and AI Agents Engineer roles, with relocation to Germany, the Netherlands, the UK or the UAE (visa sponsorship required). His notice period is 2 weeks. Consulting is available on request.
What has Huzaifa Tahir built?
Highlights: a payroll automation system that cut monthly processing time by 90% for 800+ employees; a legal AI platform at SyntaxBrain for a UK law firm with 15+ lawyers, combining RAG over 10M+ documents with Llama 3.1 fine-tuning; a CrewAI and FastAPI deep research sales agent; MCP servers for OpenAI; and a GPT-4 Vision vehicle detection model at 74% accuracy.
Which technologies does Huzaifa Tahir work with?
Python, LangGraph, CrewAI, LangChain, LlamaIndex, MCP, OpenAI, Anthropic Claude, Llama 3.1, Qwen, Whisper, FastAPI, Django, PostgreSQL, pgvector, Pinecone, ChromaDB, Docker and AWS.
How can I contact Huzaifa Tahir?
Email huzaifatahir7524@gmail.com, book a 20-minute call at calendly.com/huzaifatahir7524/new-meeting, or connect on LinkedIn at linkedin.com/in/huzaifa-tahir-ai.
Let's build something real.
I reply within a day and I'm happy to do a technical interview or a take-home task.
huzaifatahir7524@gmail.com