Akash Chauhan

Turns support tickets into reviewed knowledge-base drafts and cited answers.

• I built a seven-queue pipeline for ticket embedding, clustering, drafting, verification, publishing, RAG deflection, and gap detection. Socket.IO streams updates to the interface.
• Draft verification uses three temperature-zero judgments and up to two revision rounds before publication. The support widget returns cited answers, filters PII, and creates tickets when needed.
• Results: 46 integration tests reported passing; 0 SSN/card violations in a 784-record audit; 20% holdout set in the draft-and-verify workflow. These figures describe the project’s reported tests and audit, not a guarantee for every input.

Stack: React, Node.js, Express, MongoDB Atlas, Redis, BullMQ, Socket.IO, OpenAI API

Source on GitHub: github.com/Akashrana1001/Closed_loop

A VS Code chat extension with five AI providers and local inference.

• I built a shared provider layer with streaming chat, prompt compression, token tracking, and conversation history inside VS Code.
• Supports Anthropic, OpenAI, Gemini, GitHub Copilot, and Ollama. Credentials use the VS Code Secrets API.
• Results: 600+ marketplace downloads; 5 AI providers; Local inference through Ollama. Download count reported in my resume; it is not a live usage counter.

Stack: TypeScript, VS Code Extension API, esbuild, LLM APIs, Ollama

Open VSX listing: open-vsx.org/extension/akashchauhan1001/ayu-tokpress

Retrieves document evidence while keeping sustainability calculations in Python.

• I built a document retrieval pipeline with Titan Embeddings v2 and FAISS. The language model extracts information; Python calculates GHG, water, and POSH metrics.
• The retrieval setup uses 500-character chunks, 50-character overlap, and top-five retrieval. Model-generated text is kept separate from deterministic calculations.
• Results: 8 BRSR categories evaluated; 1.0 context precision; 1.0 context recall. Scores are from the documented local benchmark, not a claim of general accuracy.

Stack: React, Node.js, Python, FastAPI, AWS Bedrock, FAISS, BullMQ, AWS S3

Source on GitHub: github.com/Akashrana1001/Green_ledger-AI

• Built a RAG pipeline with LangChain and AWS Bedrock, exposed through Node.js REST APIs.
• Implemented a custom ReAct loop with structured outputs and tool calling.
• Evaluated 50+ queries with RAGAS to identify retrieval and prompt failures.

B.Tech, Computer Science & Engineering

• National Public School — Intermediate, 79% · August 2023
• National Public School — Matriculation, 85% · August 2021