Skip to content
View SandeepKr24's full-sized avatar
💻
Looking for internships
💻
Looking for internships

Block or report SandeepKr24

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
SandeepKr24/README.md

Hola, Waving hand I'm Sandeep Kumar

AI/ML Engineer | RAG & Agentic AI | LLMs

About Me

  • Masters in Data Science from Monash University (Distinction), and B.Tech in Information Technology from Manipal University Jaipur.
  • Focused on LLMs and Generative AI: RAG pipelines, LLM fine-tuning, and agentic AI systems, with a particular interest in AI agents.
  • Built PitWall: An end-to-end natural-language F1 analytics agent, and FieldMedic: A RAG-based clinical decision support system.
  • Co-authored a review paper on AI in solar energy processes, published at IEEE I2CT 2022.
  • Currently seeking full-time opportunities building production-grade AI systems.

Current Focus

  • Designing and hardening agentic AI pipelines: planning, tool-calling, and failure recovery.
  • Building RAG systems with vector search (pgvector, FAISS, Pinecone) and structured LLM reasoning.
  • Backend engineering with FastAPI and PostgreSQL for AI-powered applications.
  • Exploring MCP servers and multi-agent orchestration (CrewAI).

Featured Projects

End-to-end natural-language F1 analytics pipeline (Python, FastF1, PostgreSQL, LLM). Ingests 1.4M+ telemetry rows per race, secures an LLM-to-SQL agent with sqlglot AST validation and per-session Postgres Row-Level Security, and includes typed error handling, token cost caps, and structured logging.

RAG-based clinical triage system (FastAPI, Streamlit, PostgreSQL/pgvector, Gemini). Semantic retrieval over 300+ medical knowledge chunks using Sentence Transformers, combined with an AI agent pipeline for structured triage recommendations. Containerized with Docker Compose across 3 services and 4 REST API endpoints.

⚡ Agentic AI Workflow for Energy Consumption Analytics

LangChain-based agent for querying and analyzing energy consumption data, with tool-calling for dynamic SQL execution and anomaly detection. Improved task completion accuracy by 25% and cut manual analysis effort by 60% through automated reporting.

Publications

A Comprehensive Survey on the Role of Artificial Intelligence in Solar Energy Processes — 2022 IEEE 7th International Conference for Convergence in Technology (I2CT)

Reach Me


Tech Stack

Programming

Python SQL

Machine Learning

NLP Computer Vision Deep Learning Reinforcement Learning

Gen AI

LangChain Prompt Engineering RAG Agentic AI MCP Server CrewAI

Backend

FastAPI REST APIs

Databases & Cloud

PostgreSQL MongoDB MySQL FAISS Pinecone AWS Docker Git Vercel


Profile views

Tools I Use

vscode github

Keywords

AI Engineer LLM Engineer Agentic AI RAG Prompt Engineering Data Science Python FastAPI PostgreSQL Machine Learning

Pinned Loading

  1. ForzaHelper ForzaHelper Public

    Find your typa car.

    HTML 1

  2. Pits Pits Public

    AI powered F1 race analyst.

    Python 2

  3. FieldMedic FieldMedic Public

    AI powered patient triage assistant.

    Python 1