About

I'm Yash Gadodia, a product manager who builds and ships AI products. I came from engineering: five years writing code before that, on platform systems at Ninja Van (notification infra, voice calling, parcel booking) and an NLP/RAG platform at Synthesis Partners for Fortune 500 clients. I moved into product because the most interesting decisions in AI products are not implementation decisions, they are decisions about what the agent should refuse to do, how you measure whether it is doing the right thing, and which job description it is actually filling.

Today I'm the founding PM at Voltade, where we ship applied AI into production at scale. I drive strategy and go-to-market for three platform products: Envoy (a conversational CRM that breaks the dashboard convention by living in WhatsApp); Vobase (an app framework for AI coding agents, born from learning why our first no-code builder didn't work); and Volty (a self-serve multi-tenant platform, in build, pooling everything we have learnt). These serve 100+ SME accounts running 230K+ AI interactions a day at 99.65% task success rate. I am closest to the user when I am on a customer's WhatsApp watching the agent answer Kai from the bakery.

I also build personally. I hand-build customer proof-of-concepts with Claude Code; they have closed 30+ SME clients. I run seven agents on a Mac Mini in my living room, each in production. I built Parallax (a couples iOS app, React Native and Supabase, end to end) and claude-init (an open-source CLI that makes any repo AI-native in one command).

#How I build

Every line of this site, my customer proof-of-concepts, and my personal agent fleet go through Claude Code. It is not a tool I tried once; the usage dashboard shows the daily workflow, live. That hands-on contact with the tool is where my conviction about building with AI actually comes from. I know where agents fail because mine fail first: wrong tool calls, silent cost runaways, confident answers to the wrong question. I've audited my own usage to find where the real leverage is, and taught Claude how to write like me so drafts stop sounding like a press release.

The pattern is discovery-to-deployment: rapid iteration with real user feedback, from first spike to production. Evaluation fits into this as the craft that keeps it trustworthy. If you cannot measure whether your agent is doing the right thing, you have not finished shipping the product. I built WIMAUT after an OpenClaw cron job silently burned $300 because no one was watching. Since then, every agent we ship has a three-layer eval harness, a failure taxonomy, and a drift detector before it is allowed to talk to real customers. I wrote about how we do this in more detail.

#What I'm optimising for

I came into product because the best products solve jobs that scale. I want to work where that happens hands-on: forward-deployed, building applied AI into production at scale with paying customers. The shape of the role matters more than the title. Founding PM, agent PM, model behaviour PM, forward-deployed applied AI. Anywhere a wrong answer still has a name and a phone number attached.

#Now

What I'm running today, from personal agents (Claudia, Lawrence) to Voltade products (Envoy, Vobase, Volty), is on Projects. Earlier: platform engineering at Ninja Van (100M+ events/day) and co-founded AfterClass (50K+ users, still active).

#Elsewhere on the site

  • Frameworks: the named patterns I keep using when shipping AI products. Eval, behaviour spec, scope-and-state, self-learning loop, cost tiering.
  • Stack: the hardware, software, and agent infrastructure I run daily.

#Contact

If you want to talk, reach me at pirsquare.yash@gmail.com. I read everything.