Headline

Lead AI Engineer | GenAI | Guy working in AI 🔄 Web3

About

AI Engineer building production LLM systems, real-time AI pipelines, and multimodal applications across SaaS and voice interfaces. Experienced in designing and deploying agent-based systems, retrieval-augmented generation (RAG), and low-latency inference workflows in production environments. Previously worked on data platforms and backend systems handling large-scale event pipelines, which now inform my approach to building reliable, scalable AI systems end-to-end. My current focus areas: • LLM systems & agents: RAG pipelines, tool-use (function-calling) agents, evaluation-driven workflows • Inference & systems: low-latency serving, streaming pipelines, real-time orchestration • Backend & infra: Python, FastAPI, Node.js, PostgreSQL, Redis, Docker, Kubernetes, cloud deployment I focus on shipping AI systems that work reliably in production — from data and orchestration to deployment and monitoring. Open to AI Engineer / Applied AI / Inference Engineering roles where I can build and scale real-world AI systems.

Location

Guwahati

AI stack

codex

claudecode

Work experience

Lead AI Engineer at Stealth AI Startup, Sep 2025 - Present · 1 yr

Artificial Intelligence Engineer at AllysAI, Nov 2024 - Present · 1 yr 10 mos, • Led development of a Whatsapp native multi-tenant agentic platform with real-time tool execution and tenant-level isolation. • Built a legal-tech AI SaaS using LangGraph agents over 760+ structured law documents (GDPR-compliant deployment on Azure France Central) • Built a multimodal AI avatar with real-time vision (gesture/emotion detection) and sub-4s voice latency; demoed at Dubai AI Summit 2026. • Delivered multiple production AI systems (voice agents, RAG pipelines, analytics), owning architecture, deployment, and scaling.

Founder at Sarathi Studio, Jan 2024 - Present · 2 yrs 8 mos, • Built and deployed AI-driven systems across voice automation and web platforms, focusing on real-world production use cases. • Built and deployed Sarathi Voice Agent — multilingual government voice assistant (Assamese, Bodo, Hindi) with on-device STT/TTS.

Founding ML Engineer at Fundora, Jan 2025 - Dec 2025 · 1 yr

Founding Engineer – Web3 Community Infrastructure at Magic Square, May 2022 - Oct 2023 · 1 yr 6 mos, Isreal, At Magic Square, I architected the technical backbone of a Web3-native community platform from the ground up. As part of the founding team, I built and scaled on-chain community infrastructure using Discord bots, Web3 auth flows, role-based access controls (Collab.Land, Gnosis Safe), and custom automation scripts for Telegram and Twitter. I engineered growth tooling, integrated APIs for user analytics and wallet-linked engagement, and helped design the technical pipeline for MagicID and the Magic Store. Spearheaded community growth from <10k to 190k+ users and played a key role in securing a Binance grant by demonstrating scalable, data-driven user acquisition strategies.

Data Engineer at Magic Square, May 2022 - Oct 2023 · 1 yr 6 mos, • Built real-time analytics pipelines processing 90K+ user events, improving data freshness and reducing release cycles by 40%. • Developed backend systems for event ingestion, processing, and dashboarding (Pub/Sub → BigQuery). • Worked on data-driven features for user insights and retention analysis.

Education

Master of Technology, Artificial Intelligence, Tezpur University

B.Tech, Computer Science, Central Institute of Technology, Kokrajhar