Skip to content
View garg-tejas's full-sized avatar
3 followers · 4 following
  • Nagpur

Achievements

Achievement: YOLOAchievement: Pull SharkAchievement: Pair Extraordinaire

Achievements

Achievement: YOLOAchievement: Pull SharkAchievement: Pair Extraordinaire

Highlights

  • Pro

Block or report garg-tejas

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
garg-tejas/README.md

Hey, I'm Tejas 👋

Typing SVG

Pre-final year CSE (AI/ML) at IIIT Nagpur. I build AI systems that hold up outside the notebook: real constraints, real data, real failure modes.

Right now I'm deep into diffusion models, explainable AI, and agentic systems.

What I've been building

Research:

  • XAI framework for diffusion-based medical classifiers. Interpretability across 1000 denoising timesteps (repo)
  • DiffMIC-v2 from scratch. Diabetic retinopathy grading with dual-granularity conditioning, 84.1% on APTOS-2019 (repo)
  • Teaching Mistral-7B to reason with GRPO. +8% on GSM8K over SFT (experiment)

Systems:

  • Real-time PPE compliance monitoring. Event-driven, EMA fusion, stable under occlusion (SentinelVision)
  • Multi-agent research assistant. LangGraph pipelines over arXiv + Semantic Scholar (ArXiv Scout)
  • NL→SQL with chain-of-thought reasoning, 78% execution accuracy on Spider (NL2SQL)
  • Medical report parser with dual OCR + Gemini pipeline (BloodParser)

For myself:

  • Knowledge curation tool with contradiction detection. Because saving tabs isn't the same as learning (Noesis)

Stack

Languages
Python TypeScript JavaScript

ML/AI
PyTorch Hugging Face OpenCV scikit-learn

Frameworks & Tools
FastAPI Next.js React Docker PostgreSQL

What I care about

  • Interpretability, not just accuracy
  • Systems that handle real constraints: occlusion, latency, limited compute
  • Honest reporting, including the parts that didn't work
  • READMEs that actually explain the tradeoffs

tejas@tejasgarg.in • Portfolio • LinkedIn

📍 Nagpur • IIIT Nagpur '27

Pinned Loading

  1. Synthetix-CS Synthetix-CS Public

    A full-stack CS interview prep app for OS, DBMS, and CN. It turns textbook chunks into interview-style question cards, grades answers with an LLM pipeline, and serves them via spaced repetition wit…

    Python

  2. xai-diffusion-based-models xai-diffusion-based-models Public

    Explainable AI framework for diffusion-based medical image classification. Temporal, causal, and counterfactual interpretability for the DiffMIC-v2 architecture.

    Python

  3. lab-safety-monitor lab-safety-monitor Public

    Real-time AI system for PPE compliance monitoring using multi-person tracking, face recognition, and a hybrid YOLO–SAM pipeline with temporal filtering and event deduplication.

    Python

  4. teaching-mistral-to-think teaching-mistral-to-think Public

    Applied Group Relative Policy Optimization (GRPO) to induce structured reasoning in Mistral-7B, achieving improved GSM8K math accuracy through RL.

    Jupyter Notebook

  5. arxiv-scout arxiv-scout Public

    A research assistant that turns a vague topic into a curated paper set and an optional structured survey. Handles discovery, per-paper analysis, and citation graph construction across arXiv and Sem…

    Python

  6. noesis noesis Public

    AI-powered knowledge distillation app that extracts key ideas from saved content, auto-tags them, and detects contradictions across notes.

    TypeScript