~/about.md
I'm a fourth-year B.Tech student in Artificial Intelligence & Data Science at RIET, Jaipur, currently in my seventh semester. Most of what I know outside the classroom I picked up by building things — an offline RAG evaluation harness, an MCP server that gives AI agents a persistent Jupyter kernel to work in, and a diff tool that catches data quality drift before it breaks a pipeline.
I've backed that up with structured programs too — an AI Strategy & Business Intelligence internship through IBM SkillsBuild, and a hands-on Agentic AI course where I built a task-oriented agent on Botpress from scratch.
Right now I'm looking for an internship or entry-level role where I can keep building tools that solve real problems, not just tick boxes on a syllabus.
~/experience/
IBM SkillsBuild — via CSRBOX & AICTE
IBM SkillsBuild — via CSRBOX
← scroll for more →
~/skills.json
Languages
AI & Machine Learning
Libraries & Frameworks
Developer Tools & Platforms
~/projects/
01 · evaluation harness
A local, offline evaluation harness for RAG pipelines. Computes precision@k, recall@k, MRR, and hit rate, with a BM25 baseline retriever built in — no API keys or GPU required. Includes a groundedness check that flags generated answers not well-supported by retrieved context.
02 · MCP server
An MCP server that gives AI agents a persistent, stateful Jupyter kernel instead of one-shot sandboxed execution. Ships with diagnostic tools to inspect in-scope variables, profile pandas DataFrames, and validate SQL query plans before execution.
03 · CLI tool
A pandas-native diff tool that detects data quality drift between two DataFrames — schema changes, new nulls, distribution shift (PSI), and cardinality changes. Ships as a CLI supporting CSV/Parquet/JSON, with a CI-friendly exit-code mode and a test suite.
~/certifications
~/contact.sh
Building something interesting? Let's talk.