AI Engineer and Full-Stack Developer. I build LLM-powered systems, multi-agent workflows, and scalable backend architecture — the kind that actually ships to production, not just lives in a notebook.
Over the past couple of years I've worked across research labs, startups, and enterprise environments, shipping intelligent systems end-to-end. I care about reliability, observability, and building things that work at scale.
-
Lead Software Engineer @ Learnlynk — Building a multi-tenant SaaS platform with AI-powered admissions workflows, lead scoring, and CRM automation. Shipping 50+ edge functions, security hardening across the stack, and database-driven architecture migrations.
-
Contributing to Traction AI — Working on the next generation of AI-driven tooling.
-
Previously — Built media automation pipelines with multi-agent orchestration at Innerbhakti, developed enterprise RAG systems at Dover, and engineered a multimodal conversational AI assistant (VISU-Reborn) at VIT-AP's AI & Robotics Lab.
VISU-Reborn — A multimodal conversational AI assistant integrating real-time speech, vision, emotion modeling, and facial recognition. Sub-2s dialogue latency with adaptive prompt strategies, function orchestration, and persistent memory.
LUMA — An agent-based pipeline system that executes planning, tool usage, and generation steps autonomously. Built with distributed workers, external API integration, and structured outputs for reliable multi-step execution.
Languages
AI & Agents
Backend & APIs
Databases
Cloud & DevOps
Tools & Platforms
