Skip to content
View code-with-idrees's full-sized avatar
50 followers · 109 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Block or report code-with-idrees

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 is supported. 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
code-with-idrees/README.md

Typing SVG

LinkedIn Gmail Medium Kaggle LeetCode Stack Overflow

profile views

❯ whoami

researcher = {
    "name"        : "Muhammad Idrees",
    "affiliation" : "FAST-NUCES Islamabad · CS Undergraduate",
    "location"    : "Rawalpindi, Pakistan 🇵🇰",
    "focus"       : ["Machine Learning", "HPC / CUDA", "Generative AI", "NLP"],
    "mission"     : "Build AI systems that are efficient, scalable, and universally accessible",
    "status"      : "Open to Research Collaborations · PhD Opportunities · Internships",
    "contact"     : "muhammad.idrees2k25@gmail.com",
}

I'm an undergraduate AI/ML researcher working at the intersection of deep learning, high-performance computing, and generative AI. My work spans from low-level CUDA kernel optimization to building multi-agent LLM pipelines. I care about making models that actually work — fast, interpretable, and deployable.

❯ Research & Interests

🧠 Machine Learning & NLP

  • Transformer architectures (BERT, T5, encoder-decoder)
  • Sequence-to-sequence & low-resource NMT
  • Multi-agent LLM orchestration
  • Efficient fine-tuning: LoRA, QLoRA, PEFT

🖥️ High-Performance Computing

  • CUDA programming & GPU-accelerated ML
  • Parallel & distributed training (MPI, OpenMP)
  • Memory-efficient inference & FlashAttention
  • Model compression & quantization

🤖 Generative AI

  • Large language models (Gemini, Groq, Ollama)
  • Diffusion models & image synthesis
  • Autonomous AI agents & tool-use pipelines
  • Prompt engineering & RLHF

🔬 Currently Exploring

  • Sparse Mixture-of-Experts architectures
  • FlashAttention & memory-efficient transformers
  • Autonomous multi-agent reasoning systems
  • Foundation model pre-training at scale

❯ Selected Projects

📖 Intelligent RC & Quiz Generation

PyTorch BERT T5 NLP

Hybrid ML pipeline using BERT + T5 transformers to analyze text and auto-generate reading comprehension questions and quizzes. Designed as a full AI-powered educational tool.

🖼️ Denoising Autoencoder — CIFAR-10

PyTorch Deep Learning Research

Designed a 182K-parameter DAE achieving 24.62 dB PSNR and 0.8225 SSIM. Accompanied by a full 10-page LNCS-format research report with 45+ ablation experiments.

🎬 Multi-Agent AI Video Generation

Python Gemini API Groq LLMs

Architected a modular multi-agent pipeline for AI video generation. Specialized LLM agents handle scriptwriting, scene planning, and synthesis — with Gemini primary and Groq fallback.

🤝 Google Meet AI Attendance Agent

Faster-Whisper Ollama Playwright

Fully offline autonomous agent that joins live Google Meet sessions, transcribes speech, marks attendance, answers questions, and auto-generates structured PDF notes. Zero API keys.

🌐 English–Urdu Neural Machine Translation

PyTorch RNN Seq2Seq NLP

Vanilla RNN encoder-decoder for low-resource Eng→Urdu NMT. Empirical study on vanishing gradients, sequence bottlenecks, and training instability in under-resourced language pairs.

⚾ Google Cloud × MLB Hackathon

GCP Python Predictive Modeling

Production-grade MLB fan engagement platform using Google Cloud AI. Built real-time predictive models and personalized content pipelines for live game enhancement.

❯ Technical Stack

Languages

Python C++ C TypeScript JavaScript

ML / Deep Learning

PyTorch HuggingFace BERT XGBoost ONNX SBERT

Generative AI & LLMs

Gemini Groq Ollama LangChain Faster-Whisper

HPC & Systems

CUDA MPI OpenMP Linux Docker

Cloud & MLOps

GCP W&B GitHub Playwright Django

❯ Highlights
🎓 Stanford Code in Place — Section Leader 2025 Selected from a global pool to mentor students in Python & computational thinking for Stanford's international CS program
🔬 CERN Beamline for Schools (BL4S) 2025 Competed in CERN's international physics competition, proposing an original experiment applying HPC and ML concepts
☁️ Google Cloud × MLB Hackathon 2025 Built a production-grade AI fan engagement platform on GCP with real-time predictive models
📄 LNCS Research Report — Denoising Autoencoder 2025 Produced a full 10-page academic paper with 45+ ablation figures on CIFAR-10 image denoising
✍️ AI Research Writing on Medium Publishing technical deep-dives on ML, NLP, and systems for a global audience

❯ GitHub Activity

 

❯ Writing

Explore research articles, tutorials, and ML deep-dives on Medium →

Machine Learning · HPC · Generative AI

Rawalpindi, Pakistan → Open to Remote Research & PhD Opportunities Worldwide


LinkedIn Email


Pinned Loading

  1. AoC_2024 AoC_2024 Public

    Forked from NathanielJS1541/AoC_2024

    Crowd-sourced solutions to the Advent of Code 2024.

    C++

  2. Google-Meet-AI-Attendence-Agent Google-Meet-AI-Attendence-Agent Public

    🤖 Autonomous AI agent that joins Google Meet classes, marks attendance, answers questions and generates PDF study notes — using Faster-Whisper, Ollama & Playwright. Zero cloud. Zero API keys. Built…

    Python 11 1

  3. everything-claude-code everything-claude-code Public

    Forked from affaan-m/everything-claude-code

    The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

    JavaScript