Vamshi Krishna Gourwar

I’m a computer science student building production-grade AI applications—especially RAG systems and automation workflows—using Python, FastAPI, and LLM APIs. I focus on turning messy inputs into reliable pipelines that perform well in real usage.

In my current Software Engineering Intern role, I developed and maintained backend services for an AI-assisted job aggregation and application tracking platform. I built and optimized scraping pipelines using Playwright and browser automation by reverse-engineering internal APIs, then improved performance for workflows handling 6,000+ job records. I also contributed to AI-assisted recommendation and automation involving job ingestion, resume matching, and structured data processing.

On the project side, I built a Job Application Tracker with a FastAPI backend and PostgreSQL that ingests job emails via Gmail OAuth 2.0, parses structured fields, and classifies them using a TF-IDF + Logistic Regression pipeline. I designed async ingestion with Pydantic validation and SQLAlchemy ORM, then validated core flows with passing integration tests. Separately, I developed and published a pip-installable Codebase RAG Assistant CLI that integrates LLM APIs with vector search and hybrid retrieval, achieving Recall@10 of 100% on a 23-question benchmark and I am constantly upgrading it so that it’ll be a production ready RAG.

My approach is practical: measure quality with metrics, the right approach, the right stack that is to be used. My idea is clear, if I build anything then that should be of any use, to me and others. It solve an actual problem and help others.