I built the LLM interface I wished existed. interface0 is for power users: cross-provider memory & chat, multi-model synthesis, sharing & team features, granular context engineering, template prompts, forking, prompt enhancing, & more Here's a demo—see 🧵 for the rest

It's been a fun project to hack on. If you want to check it out & learn more, go to I'll keep this thread updated as a build log as I ship features btw: the base UI is forked off @ibelick’s excellent open-source project Zola. Follow him!

Here's this week's build update (and all the previous ones)... striving to answer the eternal question "what did you get done this week?"

This week's update: email as a first-class citizen.

This week: Artifacts! Collaborate on docs with AI. Like Cursor for text in your main LLM interface.

This week: mostly de-janking and speeding things up. Plus more [email protected] and Team Plans...

This week: WhatsApp integration almost shipped, turbo models, and more quality-of-life improvements.

This week: WhatsApp is live! It's great. WhatsApp voice note -> fully-tooled, fully-contextualized AI response is my new favorite flow. Plus deeplinks for Raycast and more.

This week: workspaces to keep team & personal chats and artifacts separate. Plus a bunch of fixes & polish, and more customizability.

This week: lots of prep for Google Drive integration, external security assessment, and context preview.

This week: Google Drive live — both including Drive files in Knowledge folders, plus semantically searching them.

This week: Deep Research. Plus context management improvements, Sonnet 4.5, and Artifacts on pause.

This week: parallelization & speed boosts, @triggerdotdev integration for background tasks (it's amazing), full Google Drive indexing, and more.

This week: the Google Docs AI integration that Google should have shipped themselves! Tag interface0 in a comment and it'll reply and edit the doc in suggestion mode For more on this feature & comparison videos:

Memory across all models... game changer

Nice! Looks similar to what we've built at Embra -- our main missing feature is true forking (we don't support pull request style changes). I think your design around granular context is great. A lot of scaled-RAG folks are realizing user-control matters quite a lot now.

embra looks amazing! and yes context is king — bunch of interface0's upcoming features are around context engineering

@pavansethi_ should have more than 16 likes

@pavansethi_ happy just building for myself :)

This is what it looks like when builders actually use the tools they’re building

Stoked to try it out

@saulodigital Wow 😳

This is super nifty!

This is what it looks like when builders actually use the tools they’re building

Looks like a serious upgrade for anyone deep in AI workflows Excited to try this out

awesome work! Looks like we faced a lot of the same painpoints over at Yoink AI re: current ai chatbots. Slightly different use case for us (writing), but excited to see lots of folks building to solve for current AI friction!

Great to see innovative projects like interface0 pushing the boundaries of AI interaction! It's exciting how tools like this can enhance productivity and collaboration. Looking forward to seeing how it evolves! 🚀

Exciting to see the development of interface0! The features you've outlined sound incredibly useful for productivity and collaboration. Innovation like this could be a game-changer in enhancing user experiences across the board. Keep up the great work!

Exciting work on interface0! The ability to integrate multiple AI functionalities and create a cohesive experience sounds game-changing. Remember, as we innovate in tech, understanding macroeconomic trends, including deflation and inflation shifts, also matters for long-term st

This sounds like a powerful tool for optimizing interactions! In the volatile world of crypto, having efficient interfaces is key. Looking forward to seeing the demo and how it can enhance user experience in Web3!

Exciting developments with interface0! The focus on team collaboration and context is crucial for improving productivity in today's complex workflows. This could really change the way we interact with AI. Keep pushing those boundaries! 🚀

This sounds like an exciting development in LLM interfaces! The features you've included for power users could really enhance productivity and collaboration. Can't wait to see the demo! What inspired you to focus on multi-model synthesis?

Built by someone who gets it. Legend 🙌