Divyanshu Bhargava
Founder Stac, Google Developer Expert (Flutter & Dart)
Abu Dhabi, United Arab Emirates
Actions
Divyanshu Bhargava is a Founder of Stac. He is a Google Developer Expert in Flutter & Dart and an Organizer of Flutter Abu Dhabi & Dubai. His passion for building apps took him to mobile development. These days, you’ll find him talking about Flutter and the future of tech. He is an open-source enthusiast and an avid tech blogger. When he is not coding, you can find him kayaking, reading, traveling, or thinking about coding.
Area of Expertise
Topics
Ship UI Without Shipping the App: Server-Driven UI in Flutter
Changing one screen in a mobile app takes longer than it should. You make the change, wait for app review, then wait again for users to update. Some never will. Meanwhile the web team has shipped three times before lunch.
Server-driven UI flips this. The server describes the screen, and the app renders it with real native widgets, so new flows, A/B tests and layout fixes go live without a release. It's also the idea underneath generative UI: when an AI agent builds a screen, it's using the same pattern.
I maintain Stac, an open-source server-driven UI framework for Flutter. In this talk, I'll cover:
- Building a server-driven screen with Stac, from JSON to native widgets
- Keeping six-month-old app versions working when the server changes
- Actions, state and navigation that survive a round trip to the server
- Testing and performance when your UI arrives over the network
- When server-driven UI is the wrong tool for the job
One Founder, a Team of Agents: From Idea to Paying Users
Coding agents made writing software cheap. Building a product people pay for is still hard, because code was never the only job. Someone still has to design the screens, test the edge cases, write the launch posts and convince a stranger to pay.
I've built 10+ products, mostly with Claude Code. Along the way I stopped treating it as one assistant and set it up like a small team: one setup each for design, development, testing and marketing.
In this talk, I'll show my setup for each stage:
- Design: turning a rough idea into screens and a written plan an agent can build from
- Develop: the rules and context that keep an agent building what you actually meant
- Test: getting the agent to check its own work before it says "done"
- Market: teaching an agent to write in your voice, not LinkedIn voice
- Sell: pricing, and the calls no agent can make for you
- What it really costs a solo founder, in money and in time
Guardrails for AI Agents: Secure Tool Calling in Flutter
Tool calling turns a chatbot into an agent. That's the point, and the risk. A chatbot that gets it wrong gives you a bad answer. An agent that gets it wrong sends money to the wrong person, and there's no undo.
The good news: the model never runs anything itself. It only proposes a call. Your app decides. That's where guardrails belong, not in the prompt, because the right input can talk a model out of any instruction.
In this talk, we'll cover:
- Treating every argument the model sends as untrusted input, and validating it in Dart
- Sorting actions by risk: read-only runs freely, reversible gets an undo, irreversible waits for approval
- Confirmation UI built from the real arguments, so the card can't say $12 when the call says $1,200
- Prompt injection hiding in places you don't expect, like a transaction memo or a task note
- Keeping API keys out of your app, setting limits, and logging every action so you can answer "why did it do that?"
From Wall of Text to Widgets: Generative UI with Flutter and A2UI
Ask most AI assistants to plan a weekend trip and you get a wall of text. Dates, hotels and prices, buried in paragraphs you end up reading twice. A good human assistant would just hand you three options, a date picker, and a Book button.
That’s Generative UI. Instead of replying with text, the agent replies with UI: a comparison card, a pre-filled form, a confirmation before you pay. It describes that UI in A2UI, the open protocol behind Flutter’s GenUI SDK, and your app renders it as real native widgets. The model never writes code. It can only pick from a catalog of widgets you control, and every tap goes back to the agent so the conversation keeps moving.
We’ll cover:
- How the agent, your widget catalog, and the app talk to each other.
- What breaks in production: invalid output, slow streaming, state drifting out of sync, and agents doing things nobody approved.
- When Generative UI is the right call, and when a normal screen is still better
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top