Skip to content
View Dev-AbrarMulla's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report Dev-AbrarMulla

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

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

Report abuse
Dev-AbrarMulla/README.md

Hey, I'm Abrar Mulla 👋

Typing SVG


🧭 What I'm about

  • 🎓 Computer Science student at BITS Pilani Dubai
  • 🧠 Focused on AI/ML engineering, NLP and LLM-based applications
  • 📈 Also into software development and data analytics
  • 🌱 Learning by building, one project at a time

🚀 Featured Projects

Project What it is Stack
Low-Dimensional Subspace Optimization for Unlearnable Data Data-protection research: restricts "unlearnable" image stamps to PCA or random subspaces and measures how much protection survives (76 training runs on CIFAR-10) PyTorch, ResNet-18, PCA, Colab
Dynamic Uncertainty-Aware Attribution Team NLP project: detects hallucinations in RAG answers by comparing Mistral-7B token distributions with and without the retrieved context PyTorch, Transformers, Mistral-7B
AI Ticket Triage LLM pipeline that classifies support tickets, drafts replies and routes low-confidence or critical ones to a human review queue Python, Groq API
CrewAI Building Emergency Response Four cooperating LLM agents (HVAC, fire, police, ambulance) that respond to simulated building emergencies CrewAI, Llama 3.1, Hugging Face

🔎 Interests

machine learning · natural language processing · large language models · retrieval-augmented generation · multi-agent systems · data protection · data analytics

🧰 Tech Stack

Languages

Python Java C C++ SQL HTML5

ML & Data

PyTorch NumPy Pandas scikit-learn

LLMs & Agents

Hugging Face CrewAI Groq

Always learning. Always improving.

Pinned Loading

  1. Dynamic-Uncertainty-Aware-Attribution Dynamic-Uncertainty-Aware-Attribution Public

    Hallucination detection for RAG using token-level information gain, KL divergence and semantic entropy from dual-pass Mistral-7B inference.

    Jupyter Notebook

  2. Akshay-Arun-Iyer/RiskRadar Akshay-Arun-Iyer/RiskRadar Public

    JavaScript 1

  3. AI-Ticket-Triage AI-Ticket-Triage Public

    AI contract analyzer that finds loopholes, cross-clause exploit chains and abuse scenarios, and suggests safer rewritten clauses.

    Python

  4. CrewAI-Building-Emergency-Response CrewAI-Building-Emergency-Response Public

    Multi-agent CrewAI simulation where HVAC, fire, police and ambulance agents coordinate during building emergencies.

    Jupyter Notebook

  5. Low-Dimensional-Subspace-Optimization-for-Unlearnable-Data Low-Dimensional-Subspace-Optimization-for-Unlearnable-Data Public

    Class-wise unlearnable-data stamps restricted to k PCA or random directions (CIFAR-10, ResNet-18).

    Jupyter Notebook 1