Headline

AI Engineer · Agentic & Multimodal Systems | LangChain · RAG · OpenAI GPT-4 · FastAPI | Built GenAI Solutions @ Sash.AI | M.Tech CS @ KVGCE

About

AI Engineer with 1+ year building production-grade Generative AI systems — from multimodal pipelines and autonomous agents to cloud-deployed LLM microservices. 🔧 What I build: → End-to-end RAG pipelines (LangChain + Pinecone) for semantic search & document retrieval → Multimodal Speech AI systems (Whisper STT + Sarvam TTS + GPT-4) for real-time voice agents → Agentic interview orchestration platforms with live avatar delivery via LiveKit → Production REST APIs with FastAPI, containerised on GCP using Docker 🤖 LLM & Prompt Engineering: Hands-on with OpenAI GPT-4, advanced prompt engineering (few-shot, CoT, system prompt design), and domain-specific fine-tuning for enterprise use cases. ☁️ Cloud & Infra: GCP (primary deployment), AWS Bedrock, Docker, Microservices Architecture 📚 Currently: Pursuing M.Tech in Computer Science @ KVGCE while working full-time — diving deep into autonomous multi-agent systems and multimodal AI research. Open to roles in: AI Engineering · GenAI Development · Agentic Systems · LLM Applications 📩 vvnamuth@gmail.com

Location

Bengaluru

AI stack

langchain

langgraph

langsmith

openai

Work experience

Artificial Intelligence Engineer at Sash.Ai, Feb 2025 - Present · 1 yr 5 mos, Bengaluru, Karnataka, India, AI Engineer | Aixplora Technologies Pvt Ltd (Sash.AI) Feb 2025 – Present · Bengaluru, India Built and shipped a production-grade Multimodal HR GenAI Platform from the ground up — an autonomous, end-to-end AI system for intelligent candidate screening, real-time evaluation, and automated interview orchestration. Key contributions: 🧠 LLM & Agentic Systems • Designed multi-step LLM workflows using OpenAI GPT-4 with advanced prompt engineering (few-shot, chain-of-thought, system prompt design) for candidate analysis, contextual question generation, and automated feedback summarisation • Architected agentic pipelines that autonomously drive interview lifecycles without human intervention 🔍 RAG & Vector Search • Built Retrieval-Augmented Generation pipelines using LangChain + Pinecone vector DB for semantic document retrieval, managing embeddings, tokenisation, and feature extraction end-to-end 🎙️ Multimodal Speech AI • Engineered real-time Speech AI pipelines integrating Whisper (STT) and Sarvam TTS for voice-driven candidate interactions • Integrated LiveKit for low-latency, avatar-based interview delivery — combining STT, TTS, and LLM into a single unified multimodal system ⚙️ Backend & Deployment • Built production AI microservices and RESTful APIs using Python + FastAPI following microservices architecture principles • Deployed AI systems on GCP using Docker; hands-on experience with AWS Bedrock for cloud-native AI pipelines Stack: Python · FastAPI · OpenAI GPT-4 · LangChain · Pinecone · Whisper · Sarvam TTS · LiveKit · GCP · AWS Bedrock · Docker

Education

Master of Engineering, Computer Science, KVG College of Engineering (KVGCE), Sullia

Bachelor of Engineering - BE, Computer science and engineering, coorg institute of technology