About
I’m an AI Engineer building production-grade Voice AI, Speech and LLM systems that solve real-world business problems. Over the past 2.8+ years, I’ve worked across AI engineering, automation and backend systems, with the last 2+ years focused heavily on LLM, speech and conversational AI use cases in regulated BFSI environments. My work sits at the intersection of AI research, software engineering and business problems — taking an idea from a conversation with a founder or business team, turning it into a prototype, evaluating models and approaches, and putting a reliable system into production. What I work on • Small Language Models (SLMs) and domain-specific AI systems • Voice AI & Speech Intelligence for real-world conversations • ASR / STT / TTS pipelines and speech processing • Code-mixed Indian-language speech and WER benchmarking • LLM evaluation, benchmarking and inference optimization • RAG and semantic search systems • Agentic AI, tool calling and multi-agent orchestration • Prompt & Context Engineering • AI Audit & Conversation Intelligence • Production AI backends using Python and FastAPI Systems I’ve worked on • Enterprise outbound calling and churn management platform handling 5M+ records • AI-powered invoice processing handling approximately 10K documents/month • RAG-powered knowledge platform reducing information retrieval time by approximately 80% • Automation and analytics systems saving 100+ operational hours/month • AI audit systems for transcription, dialogue analysis, scorecard evaluation, sentiment analysis and call intelligence Technologies AI / LLM: GPT, Claude, Gemini, LLaMA, Qwen, Gemma, Mistral, Phi-3, gpt-oss Agents: LangChain, LangGraph, CrewAI, MCP, RAG, Tool Calling, Multi-Agent Orchestration Speech: Whisper, Wav2Vec2, SeamlessM4T, SpeechBrain, NVIDIA NeMo, AI4Bharat, Deepgram, AssemblyAI, ElevenLabs Engineering: Python, FastAPI, PostgreSQL, Redis, Docker, WebSockets/WebRTC, React, Next.js Vector / Data: ChromaDB, Qdrant, FAISS, Pandas, NumPy, Scikit-learn I’m particularly interested in the part of AI engineering where experimentation turns into something reliable enough for production — evaluating models, improving quality, reducing latency and cost, and designing systems that can operate at scale. That’s the direction I’m continuing to explore: Voice AI, agentic systems, LLM evaluation and intelligent automation. Always interested in building, experimenting, learning and solving difficult AI problems. 🔗 Portfolio: https://rgowtham25.netlify.app 💻 GitHub: https://github.com/rgowtham25
Work experience
Python Developer (AI Engineer) at Valuestream Business Solutions Pvt ltd, Jul 2024 - Present · 2 yrs 4 mos, Bengaluru, • Built and evaluated SLM and LLM solutions for BFSI use cases, including prompt/context engineering, fine-tuning and inference benchmarking. • Built end-to-end Voice AI and Speech Intelligence systems using ASR/STT, LLMs and TTS, including code-mixed Indian-language speech evaluation and WER benchmarking. • Developed an AI Audit System for automated transcription, dialogue analysis, scorecard evaluation, sentiment analysis and structured call summaries. • Automated processing of ~10K invoices documents/m, benchmarking models for accuracy, latency and cost and implementing model-routing strategies. • Built a RAG-powered Internal Knowledge Portal, reducing information retrieval time by approximately 80%. • Designed an Enterprise Outbound Calling & Churn Management Platform supporting operations across a 5M+ record dataset. • Built automation and analytics platforms saving 100+ operational hours/month. • Worked directly with founders and business teams from problem definition → prototyping → evaluation → production deployment. • Awarded “Outstanding Performance in Techie” (FY 2025–26) for AI engineering and automation impact.
Product Engineer (L1) at ClaySys Technologies, Jan 2023 - Jun 2023 · 6 mos, Coimbatore, • Developed and integrated UI modules and automation workflows for enterprise RPA products, working across application development, workflow automation and product engineering.
