Ramyak Jain
Building with code. Sketching what's left over.
Overview
Social Links
GitHub Contributions
Hello
- I'm Ramyak Jain — a B.Tech CSE (Data Science) student, working toward forward-deployed engineering-style roles.
- Focused on Gen AI engineering and backend systems: LangChain/LangGraph, RAG pipelines (FAISS/BM25), FastAPI, agentic pipelines.
- Outside code, I sketch in ink and shoot photography — see the creative side of this site for that half of the story.
Stack
Experience
CodeFitz Technologies
- Location
- Ghaziabad, Uttar Pradesh, India
- Location type
- (Hybrid)
- Designed and developed the company's course website, delivering a clean, user-friendly interface for learners.
- React.js
- Tailwind CSS
Education
- C
- C++
- Python
- JavaScript
- DSA
- Database Systems
- Systems Design
- Distributed Systems
- Software Engineering
- Pascal
- HTML
- CSS
- JavaScript
- Debate
- Public Speaking
- Art & Craft
Communities
CodeChef ABES ECEvent Head
09.2025 – 07.2026
One contest changed everything.
A competitive-programming and problem-solving community, running since 2018. Started as Event Coordinator (Oct 2024 – Aug 2025), managing events with 750+ participants; then Event Head (Sep 2025 – Jul 2026), organizing 5+ coding events and workshops and growing student participation by 35%. Still part of the community beyond that term.
Minerva — The Literary SocietyGraphics Head
07.2025 – 07.2026
One post led to a whole new world of designing.
ABES Engineering College's official literary society — workshops spanning graphics, debate, content writing, and poetry, and known for "Fortify," its signature inter-college debate event. Led the graphics side: visual identity and design assets across the society's events and campaigns.
Projects(4)
A production-style RAG system — point it at any folder of documents and ask questions in plain English; it answers only from your documents and cites every claim by source number.
- Hybrid retrieval fusing FAISS (semantic) and BM25 (keyword) search via Reciprocal Rank Fusion, then a cross-encoder reranker re-scores the top candidates against the question.
- Provider-agnostic LLM layer (Gemini, Mistral, OpenAI, Anthropic — swap via one .env line) with a citation-enforcing prompt, so every answer is traceable to source text.
- 23 automated tests (12 embedder unit tests, 11 pipeline integration tests).
- Python
- FastAPI
- FAISS
- BM25
- Cross-Encoder Reranking
Blog(1)
In progress
- Published on