Appears on
Articles5
Design Decisions Behind app.build, a Prompt-to-App Generator
TL;DR: We built app.build, a reliable open-source AI code generator by: limiting initial scope to CRUD web apps, using FSM-guided tree-search actors, implementing extensive validation, and encapsulating context management using an error analysis feedback loop. These architectural...
app.build Can Now Build Python Data Apps
When we started working on app.build, we knew in the longer run we wanted to build a generic agent that could build apps with different “coding stacks”. However, for our first release and initial announcement, we could only build apps written with a single fixed stack: In the rec...
Six Principles for Production AI Agents
Every now and then, people ask me: “I am new to agentic development, I’m building something, but I feel like I’m missing some tribal knowledge. Help me catch up!” I’m tempted to suggest some serious stuff like multiweek courses (e.g. by HuggingFace or Berkeley), but not everyone...
Generate Apps Locally for Free: App.build Now Supports Open Source Models
App.build now supports open weights LLMs via Ollama, LMStudio, and OpenRouter – enabling you to generate complete applications end-to-end without cloud API dependency or associated costs. Why Run App.build Locally? Zero API costs Cloud LLM APIs can become expensive fast during ex...
Why we built app.build
LLMs are fantastic at code generation. They can produce hundreds of lines of code, entire functions and components, that compile perfectly, implement complex algorithms elegantly, and even follow best practices for style and structure. They suck at full-blown software generation,...