Mukunda S

Profile

The Infrastructure Mind Behind the Agentic Era

There's a moment every engineer remembers - when they stop building features and start building the thing that makes features possible. For Mukunda Srinivasagowda, that moment happened repeatedly, each time at a bigger scale. At Runnr, it was logistics infrastructure. At Zomato, it was a platform that routed over a million food deliveries a day. At Navi, it was the lending engine beneath Sachin Bansal's fintech ambitions. And at SuperAGI, it became something more fundamental: the operating system for AI agents that don't just respond to prompts, but go out and do things.

Mukunda is the Co-Founder of SuperAGI, the Palo Alto-based AI company that attracted $10 million in Series A funding from Jan Koum's secretive Newlands VC - a firm that almost never discloses its investments. SuperAGI's open-source autonomous agent framework pulled in over 17,500 GitHub stars within months of launch, the kind of number that doesn't happen by accident. It happens when engineers recognize something that actually works.

What makes Mukunda's particular vantage point interesting is what he's carried from each previous chapter. Logistics at scale taught him that systems have to handle failure gracefully at volume - there is no controlled environment. Fintech taught him that precision matters more than velocity when real money is moving. And agentic AI, it turns out, needs both. An autonomous AI agent booking a sales call or writing production code is not a prototype. It's a commitment.

The Runnr Years: Building What Nobody Else Would

In 2015, before food delivery logistics in India had been figured out, Mukunda co-founded Runnr. He didn't join as a strategist. He sat down and wrote the code himself - designed the architecture, set the engineering standards, and grew the team from nothing to 40+ engineers. The B2B logistics platform he built became valuable enough that Zomato acquired the company in 2018.

The acquisition didn't send him to a corner office. It sent him deeper into the problem. As Associate VP of Engineering at Zomato post-acquisition, he helped take the combined logistics platform to over a million orders per day - and led the migration to microservices that made that scale sustainable. Engineers who've done migrations at that volume will understand exactly how unglamorous, and how important, that work is.

Navi: Fintech at the Ground Floor

After Zomato, Mukunda joined Navi as Head of Engineering for Lending - the fintech company being built by Flipkart co-founder Sachin Bansal. This wasn't a lateral move. Lending infrastructure is legally constrained, audit-heavy, and deeply unforgiving of the kind of "move fast" ethos that works fine when you're delivering biryani. It's a different kind of discipline.

It's also where he met Ishaan Bhola, the person who would become his co-founder at SuperAGI. That kind of working relationship - building critical financial infrastructure together under real pressure - is a more reliable co-founder compatibility test than any accelerator exercise.

SuperAGI: When the Framework Came First

When Mukunda and Ishaan launched SuperAGI in early 2023, they started with a problem statement that any developer in the agentic AI space would recognize: every available framework for building autonomous agents was, in some fundamental way, broken. They tested them all. None of them could reliably take action in the real world.

So they built their own, open-sourced it under the handle TransformerOptimus, and watched 17,500 developers star it on GitHub. The framework - primarily Python, with a GUI for agent interaction, vector database integration, memory storage, and a marketplace of toolkits spanning Twitter, Jira, GitHub, and DALL-E - became the foundation for everything SuperAGI built on top of it.

The company's theory of the market is precise. Large Language Models are exceptional at generating - content, text, code drafts. What they fail at is acting. Booking the meeting. Sending the outreach sequence. Shipping the PR. SuperAGI built what they call Large Agentic Models (LAMs) - a positioning play that argues the next evolutionary step isn't just smarter language, it's language that executes.

Products

The SuperAGI Product Suite

Open Source as Architecture, Not Strategy

SuperAGI's decision to open-source its core infrastructure - including LAMs like SAM-7B and vision model VEagle - wasn't purely a growth hack. It reflects a worldview Mukunda has carried from his engineering years: systems built in the open are systems that get stress-tested faster, by more people, under more conditions than any internal team could manage. The 17,500 stars are a validation signal, not a vanity metric.

The company followed the open-source release with a suite of commercial products that sit on top of that foundation. It's the same pattern that worked for MongoDB, Elasticsearch, and Hashicorp - give engineers the infrastructure, sell the enterprise layer. SuperAGI is betting the same logic applies to agentic AI at the application level.

The Education That Started It All

Mukunda holds a Bachelor of Engineering from P.E.S Institute of Technology, Bangalore - part of Visvesvaraya Technological University - where he studied from 2006 to 2010. PES has produced a significant number of founders and engineering leaders in the Indian tech ecosystem; Mukunda's trajectory fits the pattern while being distinctly his own.

He went from campus to Amazon, where early-career exposure to distributed systems at Amazon Prime scale gave him a playbook he's been running variations of ever since. Microservices. Infrastructure. Scale. The vocabulary hasn't changed. Only the domain has.

What Comes Next

SuperAGI's stated aim is something close to a universal AI super app for work - 25+ AI-native applications in a single platform, each powered by agents that learn from every interaction. The $49/seat plus credit-based model is designed to make it accessible at the SMB level while enterprise customers can build on top of the open-source layer.

At $9.9M ARR with 190 people - in a space that's attracting capital faster than any since the original cloud wave - SuperAGI is either very early or very well-positioned, depending on which conference panel you're at. Mukunda, characteristically, is focused on the architecture rather than the narrative.

The playbook is consistent: find the infrastructure problem nobody wants to own, build it in a way that can scale to a million of anything, and open-source enough of it that the best engineers in the world are stress-testing it before you ship it to customers. He's run that play at Runnr, at Zomato, and now at SuperAGI. The domain changes. The instinct doesn't.