About
I am an AI Platform Architect and Engineering Leader with 20+ years designing production-grade systems at the intersection of agentic AI, cloud infrastructure, and distributed architecture.
The Journey
My background spans healthcare informatics, financial systems, enterprise SaaS, and safety technology — across corporate roles at Philips Healthcare and Syncplicity by Axway, and through Inovark Technologies, the global engineering firm I founded in 2017.
The through-line across all of it: staying hands-on technical while owning the outcomes — architectural, delivery, and business.
What I Do
Agentic AI Systems: I design and build production-grade agentic AI platforms — LangGraph stateful orchestration, LlamaIndex RAG pipelines, MCP tool servers, LangSmith observability, and RAGAS evaluation. Recently shipped a full platform with verified eval results: task completion 1.00, RAG faithfulness 1.00, hallucination rate 0.10.
Cloud & Distributed Architecture: I architect cloud-native systems on AWS and Azure supporting 1M+ monthly active users — with a focus on reliability, cost efficiency, and tradeoffs that hold up at scale.
Engineering Leadership: I have scaled distributed engineering teams from 5 to 60+ across the US, Europe, and India — hiring, mentoring, and building delivery culture without losing technical depth.
Beyond the Architecture
I hold two U.S. Patents in cloud storage and analytics architecture. My work spans platforms deployed globally — from the WHO Learning Portal (50K+ users) to iSafe, an AI/IoT student safety system deployed in 50+ schools.
All my AI portfolio repos are open at github.com/manisundaram — portfolio and writeups at hellomani.com/work.
This blog is where I share what I learn — architecture decisions, agentic AI patterns, engineering tradeoffs, and occasional thoughts on finance and life.
Let's Connect
Open to AI Platform Architect, Principal AI Architect, Director of AI Engineering, and Applied AI Engineer roles in the Bay Area.