Skip to content
View Riyasharma-17's full-sized avatar
💭
Learning, building, and getting better every day
💭
Learning, building, and getting better every day

Highlights

  • Pro

Block or report Riyasharma-17

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 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
Riyasharma-17/README.md

Riya Sharma

AI Engineer (Building) · 3rd Year CCE @ Amrita Vishwa Vidyapeetham · 2028
Building GenAI applications — RAG pipelines, multi-provider LLM backends, full-stack AI tools

LinkedIn · Email · Open to AI Engineer internships & collabs


🚀 Shipped Projects

Project What it does Stack Status
CLI Multi-Provider AI Chatbot Provider-agnostic LLM backend - Groq, Gemini, OpenRouter through one adapter interface. FastAPI REST API, 10 pytest tests, streaming, JSON mode Python · FastAPI · Groq · Gemini · OpenRouter · pytest ✅ Live
AI Study Assistant RAG pipeline from scratch - no LangChain, no vector DB. PDF extraction → chunking → keyword retrieval → LLM generation Python · Gemini API · PyPDF2 ✅ Live
SpeechCraft AI Real-time speech coaching - live recognition + AI feedback + gamification for dyslexia users JavaScript · Web Speech API · Groq API ✅ Live

🧠 Technical Stack

GenAI / LLMs     RAG Pipelines · LLM APIs · Prompt Engineering · Adapter Pattern
                 Groq API · Gemini API · OpenRouter · LangChain (learning)

Backend          FastAPI · REST APIs · Pydantic · async/await · python-dotenv

Languages        Python · Java · JavaScript · SQL

Frontend         React · Tailwind CSS

Testing & CI     pytest · unittest.mock · GitHub Actions

Tools            Git · Docker (basics) · Postman · IntelliJ IDEA · VS Code

📌 What Makes These Projects Different

  • Multi-provider architecture — not hardcoded to one LLM API. Groq, Gemini, and OpenRouter behind one interface. Swap providers without touching app logic.
  • RAG without frameworks — built retrieval from scratch to understand what LangChain and LlamaIndex actually hide. Chunking, scoring, Top-K, all manual.
  • Tests that verify claims — 10 pytest tests with mocked providers. Defensive JSON parsing isn't just mentioned in the README — it's tested.
  • CLI → REST API without rewrite — same provider adapters power both. That's what separation of concerns looks like when it actually works.

📈 Currently Building Toward

  • Phase 2 — RAG with vector databases (Pinecone / ChromaDB)
  • Phase 3 — AI Agents with LangGraph
  • Streaming responses + persistent conversation history
  • Deploying full-stack AI app with live URL

📊 GitHub Stats


Pinned Loading

  1. ai-study-assistant ai-study-assistant Public

    AI-powered PDF Study Assistant built with Python, Gemini API, and RAG concepts for document-based question answering.

    Python 2 1

  2. CLI-Multi-Provider-AI-Chatbot CLI-Multi-Provider-AI-Chatbot Public

    CLI chatbot + FastAPI backend — Groq, Gemini, OpenRouter through one unified adapter interface. Conversation memory, JSON output, pytest coverage, Swagger UI.

    Python 1