Sim Studio lets you build agents with different models, tools, and custom functions. In minutes, you can deploy a research agent. Demo here - https://youtu.be/F6MqNvnfxbI

We started working on Sim Studio because we were frustrated with agent frameworks that required excessive boilerplate code and unhelpful abstractions (like agent roles and backstories). We also tried other visual interfaces, but found them difficult to set up and too limited for production use cases. Specifically, we wanted the ability to write custom functions, force tool calls, and run parallel execution.

We’re taking a different approach. Our visual canvas lets you explicitly define how agents interact, use external tools, and handle complex logic like branching, loops, and conditional execution. You can seamlessly integrate with popular tools like Supabase, Pinecone, Mem0, and Exa without obscure abstractions.

You can simulate workflow runs, deploy workflows as APIs or standalone chat interfaces, and trigger them via webhooks. We offer built-in observability, detailed trace spans, and logs that make debugging straightforward. Try searching yourself below:

Example message for https://hn.simstudio.ai:

- Search [person] from [company] - "Search Emir Karabeg from Sim Studio"

Check out our Apache 2.0 licensed repo at https://github.com/simstudioai/sim, explore our docs at https://docs.simstudio.ai, or try out our deployed chat from the demo video here: https://hn.simstudio.ai.

We’d love to hear your feedback! Do you think our visual solution is a good approach to building agents?