Skip to content
View debarnabdas007's full-sized avatar
8 followers · 33 following

Block or report debarnabdas007

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
debarnabdas007/README.md

Typing SVG


Building deployable ML systems that ship, not just demo notebooks

Focused in Applied ML, Data Science and Python Backends. Messy Data In --> Reliable Predictions Out --> Deployed APIs.
I care deeply about the fundamentals: clean feature engineering, rigorous evaluation, and shipping code that works.

⚡ Technical Arsenal

Category Technology Stack
👨‍💻 Languages
🧠 Machine Learning
🔍 Search Arch & LLMs
🔌 Backend & APIs
🎨 Frontend & UI
🗄️ Databases
🛠️ Tools & DevOps

🚀 Engineered Solutions

📂 Project Name 🛠️ Tech Stack 🔗 Links
🚦 Traffic-IQ: Edge AI Vehicle Analytics FastAPI OpenCV Scikit-learn Repo
✈️ AEGIS AERO: Predictive Maintenance System XGBoost Flask Python Repo   Live
🏙️ Sahayak: Civic Intelligence Platform Gemini API Flask Python Repo   Live
✍️ Hand-Digit Recognition Pipeline *(*interactive)(Built from Scratch) PyTorch FastAPI Streamlit Repo   Live
🧠 Agentic-RAG: Production-Grade AI Backend *(No Frameworks) FastAPI Cross-Encoder Llama 3 (Groq) FAISS/BM25 Custom State Machine Repo   Explanation

📌 Experience

GDG On Campus – Cybersecurity Wing (2024–25)

  • Core Team Member selected after review
  • Hands-on exposure to Linux CLI and core networking fundamentals
  • Explored web vulnerabilities (like SQL Injection and more..) via hands-on (Kali-VMware, TryHackMe-CTFs)
  • Applied systems-level understanding to secure API development and deployments

📊 GitHub Activity

GitHub Stats Top Languages

GitHub Streak

Profile views

" Turning raw maths into ruthless APIs⚡"

Pinned Loading

  1. Agentic-RAG- Agentic-RAG- Public

    Production Oriented Agentic-RAG-System :- LLM tool selection, Multi-step reasoning, Memory, Agentic Loop, Failure handling and proper Logging for Debugging.. Later added: session isolation, eval ha…

    Python

  2. Aircraft-Engine-RUL-prediction Aircraft-Engine-RUL-prediction Public

    Jet engine RUL prediction with 96% model size reduction (162MB --> 5.5MB) for cloud deployment. XGBoost + 42 temporal features + cold-start simulation. Live Flask API on Render

    Jupyter Notebook

  3. Learning-Pytorch Learning-Pytorch Public

    Pytorch from Fundamentals using Google Collab GPU

    Jupyter Notebook

  4. Traffic-IQ Traffic-IQ Public

    Real-time vehicle detection & classification on CPU-only edge devices. Classical CV pipeline (MOG2 + KNN) achieves 60+ FPS with no GPU. FastAPI streaming backend with MJPEG and REST endpoints. Topi…

    Python

  5. fastapi-best-practices fastapi-best-practices Public

    Forked from zhanymkanov/fastapi-best-practices

    FastAPI Best Practices and Conventions we used at our startup

  6. Hand-digit-recognition-system-from-SCRATCH Hand-digit-recognition-system-from-SCRATCH Public

    End-to-end handwritten digit recognizer built from scratch in PyTorch -- no high-level abstractions. FastAPI inference backend + interactive Streamlit frontend. Live on HuggingFace Spaces. Build on…

    Jupyter Notebook