AI / ML Engineer · Currently at Invisible Technologies · Karachi, Pakistan

AI engineer working on fine-tuning, reinforcement learning and training data.

I’m Abdul Rauf. I adapt language models to real tasks, improve them with feedback, and build the data and evaluation work that makes the results hold up.

Abdul Rauf profile photo

What I work on

Areas I’m exploring and building in

Fine-tuning

Adapting foundation models with supervised fine-tuning, LoRA and QLoRA for practical domain-specific tasks.

Reinforcement learning

Exploring preference optimization and feedback-driven methods to improve model behaviour and reliability.

Data and evaluation

Creating high-quality training data, evaluation workflows and benchmarks that make model improvements measurable.

Selected work

Projects, experiments and open-source work

PolicyPilot

An AI-powered assistant for understanding policy documents and turning complex information into useful answers.

View project ↗

AgriMatch

A data-driven platform concept connecting agricultural needs with relevant resources and recommendations.

View project ↗

Transformers

Hands-on experiments with transformer architectures, language models and model fine-tuning workflows.

View project ↗

Omnigent

Exploring generative AI systems and practical applications built around modern language models.

View project ↗

About

I’m a computer science graduate focused on AI, machine learning and large language models.

I’m interested in the full model development cycle — from preparing and curating data to fine-tuning, evaluation and deployment. I enjoy turning research ideas into working experiments and learning by building.

Tools I use

Languages

Python · C++ · TypeScript · SQL

Machine learning

PyTorch · TensorFlow · scikit-learn

LLMs

Hugging Face · OpenAI APIs · Fine-tuning · RL

Data

NumPy · Pandas · Jupyter · DVC

Databases

PostgreSQL · MySQL

Let’s talk about your next model.

I’m always interested in thoughtful conversations about language models, AI training, evaluation and opportunities to build useful things.