Hi, I'm
AI/ML Engineer | Data Scientist
Building Intelligent Systems and Turning Data into Intelligent Solutions
About Me
I’m a passionate AI Engineer specializing in Generative AI, Machine Learning, and Large Language Models, with expertise in building intelligent systems that combine cutting-edge AI techniques with practical applications. My work focuses on Retrieval-Augmented Generation (RAG), prompt engineering, and LLM evaluation to create reliable and context-aware AI solutions.
I have hands-on experience developing hybrid evaluation workflows, curating high-quality datasets for fine-tuning, and implementing robust validation pipelines. I'm particularly interested in AI safety, hallucination detection, and building trustworthy AI systems that can be deployed in real-world scenarios.
Beyond tech, I find joy in exploring emerging AI technologies, contributing to open-source projects, and collaborating with teams to solve complex problems. I’m a firm believer in responsible AI development and continuous learning in this rapidly evolving field.
Education
Bachelor of Science in Computer Science
FAST-NUCES
Lahore, Pakistan
Sept 2021 – June 2025
Experience
AI Engineer
AnalytiverseJune 2025 – Present
Lahore, Pakistan
- •Developed hybrid evaluation workflows combining LLM-as-a-Judge frameworks with manual validation to systematically detect and penalize hallucinations in agentic systems.
- •Curated high-quality preference datasets (chosen vs. rejected) and instruction-tuning data to support Supervised Fine-Tuning (SFT) and alignment tasks.
- •Performed error analysis on model outputs to identify failure modes, using these insights to refine system prompts and improve response faithfulness.
- •Built Python-based validation scripts to ensure data integrity and consistency before feeding into training pipelines.
Generative AI Intern
Systems LimitedApril 2025 – June 2025
Lahore, Pakistan
- •Contributed to an Agentic Developer Assistant tool capable of analyzing code context and suggesting improvements.
- •Integrated Retrieval-Augmented Generation (RAG) components to fetch relevant documentation, improving the accuracy of code suggestions.
- •Designed logic for automated bug detection agents that utilize AST parsing to identify syntax and logic errors.
- •Collaborated with the engineering team to optimize the response streaming pipeline for a smoother user experience.
Technical Skills
Generative AI
- RAG
- LangChain
- Prompt Engineering
- Dataset Curation (SFT/RLHF)
- LLM Evaluation
Machine Learning
- PyTorch
- TensorFlow
- Transformers
- Computer Vision
- NLP
- Scikit-learn
MLOps & Tools
- Docker
- Git
- Linux
- FastAPI
Languages
- Python
- SQL
- C++
- JavaScript
Projects
Built a RAG pipeline for legal document analysis, using DeepSeek-R1 for reasoning and FAISS for vector retrieval. Implemented prompt engineering techniques to strictly ground model answers in retrieved context, minimizing fabricated legal advice.
Designed an image restoration pipeline for degraded Urdu text, utilizing SCUNet for denoising and SwinIR for super-resolution. Created a synthetic dataset with controlled noise and blur artifacts to train models for low-resource language scripts.