aantti
Re:Deno Is Joining Cloudflare
on Saturday October 10, 2026 @06:21AM (#50031426)
Edge functions use https://github.com/supabase/edge-runtime/ (which uses an older version of Deno), and run on Supabase-managed infra :)
Re:Supabase Is Acquiring Turso
on Saturday October 03, 2026 @05:02AM (#49942536)
Re:Launch HN: Langfuse (YC W23) – OSS Tracing and Workflows to Improve...
on Tuesday December 17, 2024 @01:39PM (#42443982)
great product & great team, kudos & congrats! :)
Fabricator – OSS framework to generate datasets with LLMs
Re:HashiCorp Adopts Business Source License
on Friday August 11, 2023 @08:17AM (#37087827)
This is very sad but also not very surprising. It's notoriously hard to sell an 'open core' product (mostly, against yourself OSS) and structure the entire GTM and the org properly for scale. Having built a successful 'open core' company before & judging from the first-hand experience - this looks like an extremely desperate move, indeed. (Btw, we never changed the license, but discussed it often, ofc.) It's also all very disruptive, destructive and hostile to the community - it's 'we aren't open source anymore,' so I'm surprised they are trying to convince people otherwise (gaslighting, eh?) I'd expect the 'open core' enthusiasm of the past few years to decrease dramatically, and I wouldn't recommend an 'open core' path to anyone who's trying to build an actual big company around their OSS. I remember the time when 'open core' was a taboo word, post-MySQL/Oracle. I was doing the company launch and accidentally told a reporter we were doing [something like] open core - that didn't work nicely :) It's ironic and sad, BSL comes from the very same folks who basically invented 'open core', then spoiled it forever. It's been also always thought-provoking to me, too, those folks never build more - or differently - after MySQL.
Show HN: Haystack – Production-Ready LLM Framework
Show HN: Memory Enables Conversational Agents in Haystack
Re:Emerging Architectures for LLM Applications
on Wednesday June 21, 2023 @08:09AM (#36417434)
Thanks for the mention and being part of Haystack's community :)
Show HN: The beginner's guide to LLM prompting
What's the difference between generative and extractive models
Generative vs. Extractive Language Models
Show HN: What are prompt injections and how to prevent
Show HN: How to Prevent Prompt Injections
Show HN: How to Write better Python logs with structlog
Using Structlog for Writing Professional Python Logs
Re:RasaGPT: First headless LLM chatbot built on top of Rasa, Langchain...
on Monday May 08, 2023 @08:42AM (#35860923)
This was an interesting read :) https://www.pinecone.io/learn/javascript-chatbot/
Re:RasaGPT: First headless LLM chatbot built on top of Rasa, Langchain...
on Monday May 08, 2023 @07:48AM (#35860516)
Also, with Haystack and a smaller Transformer model to address the long-tail of answers https://github.com/deepset-ai/rasa-haystack (and https://www.deepset.ai/blog/build-smart-conversational-agent...)
Re:Re-implementing LangChain in 100 lines of code
on Friday May 05, 2023 @02:52AM (#35826068)
Yes, Haystack is Apache 2.0 :) https://haystack.deepset.ai