Skip to content
View sarthmish's full-sized avatar

Block or report sarthmish

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
sarthmish/README.md

πŸ‘‹ Hi there, I'm Sarthak Mishra!

Typing SVG

LinkedIn GitHub Email Location


πŸš€ About Me

πŸŽ“ Final-year Integrated M.Tech Student in Software Engineering at VIT Vellore
πŸ€– Currently interning as SWE-AI Intern at Bosch Global Software Technologies
πŸ”¬ Specializing in Generative AI, RAG Systems, and Scalable AI Solutions
🎯 Actively seeking full-time opportunities in AI/ML and Software Engineering

"Building the future with AI, one algorithm at a time"


πŸ’Ό Current Work at Bosch

Bosch

πŸ€– Developing GenAI-powered Calibration Chatbot using RAG, Local LLMs (Llama, DeepSeek, Gemma)
πŸ“Š Building EV Range Prediction models with MATLAB for SOC analysis
πŸ› οΈ Engineering Python-based CAN Tools for automotive calibration
πŸ“ˆ Reduced search time by 60% through intelligent document querying systems


πŸ› οΈ Tech Arsenal

Languages & Core Technologies

Python Java SQL MATLAB JavaScript

AI/ML & Data Science

TensorFlow PyTorch OpenCV scikit-learn LangChain

Web & Backend

FastAPI Django Flask Streamlit

Databases & Cloud

MongoDB MySQL Azure Docker


🎯 Featured Projects

πŸ€– Cali-Bot - GenAI Calibration Assistant

RAG β€’ Local LLM β€’ Vector Search β€’ FastAPI

Secure chatbot with 60% faster document querying, supporting multiple formats (PDF, Word, Excel, PPT)

πŸ₯ Medical Record Digitization System

Computer Vision β€’ NLP β€’ Healthcare AI

End-to-end conversion of handwritten records to FHIR format using CNN-BiLSTM architecture

πŸ›‘οΈ XAI Intrusion Detection with LEMNA

Explainable AI β€’ Cybersecurity β€’ Deep Learning

81.5% accuracy across 9 attack types with 13M+ parameter neural network

πŸ’¬ Full-Stack RAG Chat Application

RAG β€’ LLM β€’ Vector Search β€’ Full-Stack

Interactive PDF chat interface with advanced retrieval-augmented generation


πŸ“Š GitHub Analytics

Top Languages


πŸ† Achievements & Certifications

IBM AI Analyst Certification

Walmart Advanced SWE Job Simulation

UCSD Big Data Introduction


🌟 What I'm Working On

  • πŸ”­ Currently building: Advanced RAG systems and GenAI applications at Bosch
  • 🌱 Learning: Latest developments in Large Language Models and Multimodal AI
  • πŸ‘― Looking to collaborate on: Open-source AI projects and innovative ML solutions
  • πŸ’¬ Ask me about: Generative AI, RAG systems, Computer Vision, and Healthcare AI
  • ⚑ Fun fact: I love exploring the intersection of AI and real-world problem solving!


🀝 Let's Connect!

Ready to collaborate on exciting AI projects or discuss the latest in tech?

LinkedIn Email GitHub


Profile Views Followers

"Innovation distinguishes between a leader and a follower." - Steve Jobs

Pinned Loading

  1. An-Integrated-Deep-Learning-Approach-for-the-Digitization-of-Health-Records An-Integrated-Deep-Learning-Approach-for-the-Digitization-of-Health-Records Public

    Leveraged Deep Learning , NLP and Computer Vision to convert handwritten medical records to FHIR format, showcasing proficiency in advanced technologies

    Jupyter Notebook

  2. Multi-Modal-Video-Generation Multi-Modal-Video-Generation Public

    Python

  3. Voice-Automated-RAG Voice-Automated-RAG Public

    Python

  4. XAI-for-Intrusion-Detection-System-using-LEMNA XAI-for-Intrusion-Detection-System-using-LEMNA Public

    Developed explainable AI model using LEMNA to interpret intrusion detection neural network. β€’ Built deep neural network with 13M+ parameters, achieving 81.5% accuracy on intrusion benchmark. β€’ Engi…

    Jupyter Notebook

  5. Full-Stack-RAG-Chat-Application Full-Stack-RAG-Chat-Application Public

    A retrieval-augmented generation (RAG) chat interface that allows users to upload PDF documents and ask questions about their content. The system uses advanced LLMs and vector search to provide acc…

    Python

  6. Web-Scraping-LLM Web-Scraping-LLM Public

    Python 1