Skip to content
View nabeelshan78's full-sized avatar
1 follower · 7 following

Highlights

  • Pro

Block or report nabeelshan78

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
nabeelshan78/README.md

Nabeel Shan

AI/ML Engineer & Researcher | Software Engineer @ NUST ('27)

Building scalable AI systems — from foundational NumPy implementations to production-grade LLMs.

🚀 Technical Focus

  • Generative AI & LLMs/VLMs: Developing scalable intelligent systems across NLP and Multi-Modal domains.
  • Computer Vision & Reinforcement Learning: Building robust visual perception models and adaptive decision-making systems.
  • Agentic AI & RAG: Architecting autonomous agents and high-precision retrieval pipelines for complex reasoning.

🛠️ Core Competencies

I architect solutions using deep algorithmic knowledge and modern frameworks.

  • Primary Focus: Generative AI, LLMs & VLMs, NLP, Computer Vision, Reinforcement Learning, Agentic AI, RAG.
  • Tech Stack: Python C++ PyTorch TensorFlow Hugging Face Scikit-learn LangChain LlamaIndex Docker Git AWS

🎓 Experience & Education

  • B.E. Software Engineering | NUST, Islamabad (2023 - 2027)
  • Research Intern | CETQAP
  • AI/ML Intern | Software Productivity Strategists (SPS)

Pinned Loading

  1. reinforcement-learning-human-feedback-scratch reinforcement-learning-human-feedback-scratch Public

    End-to-end implementation of Reinforcement Learning with Human Feedback (RLHF) to align a GPT-2 model with human preferences — covering Supervised Fine-Tuning (SFT), Reward Modeling, and PPO-based …

    Jupyter Notebook 2

  2. safe-llm-adaptation-peft-rlhf safe-llm-adaptation-peft-rlhf Public

    An end-to-end pipeline for adapting FLAN-T5 for dialogue summarization, exploring the full spectrum of modern LLM tuning. Implements and compares Full Fine-Tuning, PEFT (LoRA), and Reinforcement Le…

    Jupyter Notebook 1

  3. Transformer-Adaptation-Playbook Transformer-Adaptation-Playbook Public

    An empirical study of Transformer adaptation techniques. Pre-training from scratch (MLM), classic fine-tuning, and from-scratch implementations of PEFT methods (LoRA, Adapters). Tuning both encoder…

    Jupyter Notebook 1

  4. gpt-forge-from-scratch-transformer gpt-forge-from-scratch-transformer Public

    A clean, modular implementation of a decoder-only Transformer (mini-GPT) from scratch in PyTorch for autoregressive text generation.

    Jupyter Notebook 2

  5. pixelsense-ai-segmentation pixelsense-ai-segmentation Public

    U-Net-based semantic segmentation on images - pixel-level road and environment understanding.

    Jupyter Notebook 1

  6. math-vlm-finetune-pipeline math-vlm-finetune-pipeline Public

    A production-ready, modular fine-tuning pipeline for converting handwritten mathematical expressions into LaTeX using Google's PaliGemma 3B and QLoRA.

    Jupyter Notebook 1