Looking for a job

Hassan Shahzad

@hassanshahzad3

Message

I build and evaluate LLM applications, from agentic RAG systems to fine-tuned Text-to-SQL models.

Pakistan

Message

What I'm looking for

I'm looking to build and evaluate practical AI applications, particularly LLM, RAG, and NLP systems, while continuing to deepen my experience in model evaluation, fine-tuning, testing, and deployment.

I've evaluated frontier LLM outputs at Turing, including Google Gemini, Anthropic Claude, and DeepSeek, assessing correctness, coding quality, and adherence to evaluation protocols.

I authored reference implementations and public and private test suites, and built pytest-based automated testing with 90%+ line and branch coverage across evaluation pipelines.

I built an agentic RAG system for Hugging Face Transformers and PEFT documentation using Python, LangGraph, ChromaDB, FastAPI, Docker, hybrid retrieval, and GitHub source-code lookup. Its evaluation reached 0.81 Faithfulness and 0.94 Answer Relevancy using RAGAS.

I've also fine-tuned Llama 3.1 8B for Text-to-SQL using QLoRA and PEFT, achieving approximately 80% execution-equivalent accuracy, and built AssignGuard, an NLP-based academic similarity detection system.

Experience

Work history, roles, and key accomplishments

LLM Evaluation Engineer

Turing

Aug 2025 - Dec 2025 (4 months)

Evaluated outputs from frontier LLMs including Google Gemini, Anthropic Claude, and DeepSeek, assessing correctness, coding quality, and adherence to evaluation protocols. Authored ideal reference implementations and designed public/private test suites to benchmark model-generated code across varied difficulty levels.

LLM Evaluation Python Pytest Testing Automation Test Suites Benchmarking Documentation