Viewing profile — spstoyanov

Thank you for the kind words!

Something important I forgot to add: AgentCouch is currently only available in US and Canada. Apologies if you wasted time trying it in other regions!

Hello HN! I wanted to share a small MCP + web app I built to solve a problem at work. Essentially, we’ve been doing great when working with agents 1:1, but collaboration has been a…

Who would have thought

So same as Sol? I guess I’ll see which one is more token efficient.

It's not only you, I am really curious about their data privacy.

Glad to see the price is going down but it's still too high for a "fast" model

The other day I was thinking about the future of programming and how our workflows change with AI and I started wondering what would an AI-native programming language look like. So…

The other day I was thinking about the future of programming and how our workflows change with AI and I started wondering what would an AI-native programming language look like. So…

Nice! Explicit is better.

cool idea!

The other day, I was wondering how big of a model I can train on a single GPU in one month. I found this great blog post and I thought it would be cool to make a simple calculator:…

Thank you so much! The main difference between langchain and LLMFlows is in the philosophy - langchain has a "chain for everything" approach where chains come with multiple LLM cal…

A few weeks ago, during the LK-99 hype, I made a demo app to explain superconductors in simple terms using RAG over relevant Wikipedia pages. I thought it would be cool to extend t…

This is awesome! Thank you for sharing!

brain activity

It's definitely not only you

Hey HN, I was dogfooding LLMFlows over the weekend and decided to build a simple FastAPI app that explains superconductors in ELI5 style.

Thank you for your feedback! And thank you for sharing guidance! You might find this guide useful to your use case: https://llmflows.readthedocs.io/en/latest/user_guide/Async%2... …

Thank you for the kind words!

This is a very hard problem. If you base your response on a text that has been returned from a vector DB you can reference this text. But if you rely on a citation generated direct…