Resume GitHub LinkedIn

[ SYSTEM ONLINE ]

SAHIL CHAVAN

AI Research Engineer

Building Intelligent Systems at the Intersection of Speech AI & Gaming

START MISSION

[ MISSION BRIEFING ]

About The Operative

AI Research Engineer focused on building the Intelligence Layer — Context Management, Memories, Prompt Orchestration, Tool Orchestration, Tool Calling, and MCP — with production-grade Prompt Engineering (System Prompt, Developer Prompts with functions & tools, Web-Search, Holding response) at its core.

Also deep in FPS AAA Gaming AI Agents — behavioral cloning, computer vision, and real-time decision systems that make bots play like humans in competitive FPS environments.

Alongside that: Speech AI systems (LID, Wake-Word, VAD) shipped in production. Bridging research and real-world agents is the mission.

CV Download Resume

[ SIGNAL MAP ]

Tap a node. Dig deeper.

[ SIGNAL LOCKED ]

Speech AI

Production LID, wake-word, and VAD systems across Indic languages — low latency, always-on, noise-robust.

  • MMS
  • ECAPA-TDNN
  • wav2vec 2.0
  • Silero VAD

[ MISSION HISTORY ]

Combat Experience

  • Fine-tuned LID models (Facebook MMS, ECAPA-TDNN) for 10 Indic languages; improved robustness under noisy, real-world audio.
  • Designed and deployed wake-word detection (wav2vec 2.0) for “Hello Jio” — low-latency, always-on inference.
  • Implemented Silero VAD with ONNX runtime and open-source pip; benchmarked accuracy and inference.
  • Architected production-ready Prompt Engineering for Gemini Live — System Prompt, Developer Prompts with functions & tools, Web-Search prompt, and Holding response prompt.
  • Built Intelligence Layer Optimising — Context Management, LLM routing, Memories, Prompt Orchestration, Tool Orchestration, Tool Calling, and MCP building.
  • Implemented prompt versioning with Langfuse & Firebase for controlled deployment, evaluation, and iteration of production prompts.
  • Improved AI agent performance by 30% through dataset optimization and algorithm tuning.
  • Reduced AI training time from 8 hours to 4.5 hours (43%) via optimized metadata and preprocessing pipelines.
  • Developed AI-driven FPS gaming agents using deep learning and behavioral cloning.
  • Applied Computer Vision (OpenCV) for real-time perception and decision-making in FPS bots.
  • Collaborated in a 4-member AI team; reduced gameplay bugs by 20% via adaptive decision logic.
  • Built metadata collection and labeling systems to accelerate experimentation cycles.
  • Supported continuous improvements in AI agent stability and realism.

[ ARSENAL ]

Project Showcase

Speech AI

Language Identification System

Fine-tuned MMS and ECAPA-TDNN models for Indic languages, addressing accent variation, background noise, and code-mixed speech.

Speech AI

Wake-Word Detection

Built a wav2vec-based wake-word system optimized for low false positives and low latency for the Hello Jio phrase.

Prompt Eng

Prompt Engineering

Production-ready prompt systems — System Prompt, Developer Prompts with functions & tools, Web-Search, and Holding response prompts.

Intelligence

Intelligence Layer Optimising

Context Management, LLM, Memories, Prompt Orchestration, Tool Orchestration, Tool Calling, MCP building.

Gaming AI

FPS Gaming AI Agents

Developed AI-driven FPS gaming agents using deep learning and behavioral cloning for realistic gameplay.

Speech AI

Voice Activity Detection

Implemented Silero VAD using ONNX runtime, benchmarking accuracy and inference performance.

[ ESTABLISH CONNECTION ]

Initialize Contact

EMAIL sahildchavan04@gmail.com PHONE +91 9834020445

Open for opportunities

· Speech AI · LID · Wake-Word · Prompt Engineering · Intelligence Layer · Computer Vision

Pune, India

GitHub LinkedIn