Skip to content
View ShounakShelke's full-sized avatar

Highlights

  • Pro

Block or report ShounakShelke

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. 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
ShounakShelke/README.md

Third Year Computer Engineering Undergraduate | Data Science and Machine Learning | Java Full Stack Developer

I build systems that don’t just predict — they act, scale, and survive real-world constraints.

Most of my work sits at the intersection of: Machine Learning | Real-time Systems | Retrieval-Augmented Intelligence (RAG) | Backend and APIs


What I Build

I focus on building systems that are practical, scalable, and deployable in real-world environments.

My work involves:

  • End-to-end machine learning pipelines: Moving beyond notebooks into production systems.
  • Computer vision and deep learning systems: Specializing in YOLOv8, object detection, and transfer learning.
  • Retrieval-augmented systems (RAG): Building smart, semantic-search-enabled systems for complex data.
  • Backend systems for model deployment: Efficient model integration using modern APIs and microservices.

How I Build

I don’t treat ML as isolated models. I build systems where Data flows → Processing → Model → API → Deployment → Feedback loop.

Animated Skills
Domain Typical Tech Stack Snippets
ML and Data PyTorch TensorFlow XGBoost Scikit-learn YOLOv8 Pandas NumPy Seaborn
CV and Deep Learning CNNs Transfer Learning Image Classification Object Detection
RAG / NLP FAISS Embeddings Sentence Transformers Semantic Search Vector DBs
Backend/APIs FastAPI Flask Spring Boot REST APIs Microservices Architecture
Databases MySQL PostgreSQL MongoDB SQLite Neo4j Apache Cassandra
Languages Java Python C R SQL JavaScript TypeScript HTML CSS PHP
Tools Git GitHub Postman Swagger Jupyter Google Cloud Figma QGIS

What I Optimize For

  • Speed and execution: Building systems that handle high throughput.
  • Practical and deployable solutions: Ensuring code works beyond a local environment.
  • Integration between ML and backend systems: Closing the gap between prediction and action.
  • Reliability in real-world scenarios: Resilient architecture under pressure.

Philosophy

Most ML projects stop at prediction.
I focus on what happens after.


Find Me


If you're building something hard

I’m interested in:

  • Real-world AI systems: Systems that solve tangible, complex problems.
  • High-speed decision systems: Where every millisecond count.
  • Applied ML: Moving beyond experimentation into value delivery.

Let’s build.


Pinned Loading

  1. Project-JV1-GP-Stratz Project-JV1-GP-Stratz Public

    A deterministic motorsport strategy environment for evaluating AI agents. Built for the OpenEnv Hackathon using FastAPI, Docker, and LLM-based inference. Features real-time race physics, tyre degra…

    Python

  2. Project-LMH-Lab-Lens Project-LMH-Lab-Lens Public

    Lab-Lens is an AI-powered medical report interpreter that converts complex lab values into simple, human-friendly insights using OCR, Gemini, and rule-based medical logic. It provides clarity, seve…

    TypeScript 2

  3. Project-2H4E-Endurance-Dashboard Project-2H4E-Endurance-Dashboard Public

    Project 2H4E is an AI-powered endurance racing dashboard for live timing analytics, telemetry-style strategy, race commentary intelligence, circuit reporting, and race-engineering insights across N…

    TypeScript

  4. Project-WRT-2-Object-Detection Project-WRT-2-Object-Detection Public

    Jupyter Notebook

  5. Project-McCaren-MedMatrix-Application Project-McCaren-MedMatrix-Application Public

    Project McCaren is an AI-powered healthcare platform for medical OCR, fraud detection, smart claim verification, predictive analytics, and intelligent healthcare assistance using multi-agent AI/ML …

    TypeScript

  6. AAI-Chatbot AAI-Chatbot Public

    HAL AEROTHON 2026

    JavaScript