Emergent, the AI coding startup co-founded by twin brothers Mukund and Madhav Jha, just became India's newest unicorn: a 130 million dollar Series C led by Creaegis at a 1.5 billion dollar valuation, about a year after it launched publicly, taking its total funding to 230 million dollars. Its roughly 200-person team is based mostly in Bengaluru. Read the announcement. The pitch is simple: describe what you want built in plain English, and a set of autonomous agents design the interface, wire up the backend and ship a production-ready app, no code required from you. Its own founder has said most of the people actually building on the platform have never written a line of code in their life.
I want to be fair to what that means before I react to it. That is a real unlock. The gap between having a product idea and having something a user can touch used to be measured in months and a hiring budget. Emergent is betting that gap collapses to a conversation.
But I build this way every day, directing AI rather than writing the code myself, on InfiNotes and on an unlaunched AI study companion, and here is what a 1.5 billion dollar valuation does not change: someone still has to know when the agent is wrong. Not wrong in syntax, agents mostly get that right now. Wrong in judgment: the feature nobody asked for, the flow that optimises for a metric instead of a person, the screen that looks finished and is not. Removing code as the bottleneck does not remove the bottleneck. It moves it, from can you build this, to do you actually know what to build, and when to tell the agent no.
The honest caveat: I have not used Emergent myself, so I cannot tell you if its review and testing layer is good enough to catch what a human founder would catch before shipping. That is the part worth watching. The valuation is not.