efriis's comments

On: Show HN: Velvet – Store OpenAI requests in your ow...

[efriis]

Howdy Erick from LangChain here! Just a quick clarification that LangSmith is designed to work great for folks not using LangChain as well :)

Check out our quickstart for an example of what that looks like! https://docs.smith.langchain.com/

[turnsout → efriis]

TIL! LangSmith is great.

On: Ask HN: How do you deploy Python-based LangChain o...

[efriis]

Erick from LangChain here - would recommend either deploying with LangGraph Cloud or anywhere that deploys a FastAPI easily.

LangGraph Cloud: https://langchain-ai.github.io/langgraph/cloud/ Example (chat-langchain): https://github.com/langchain-ai/chat-langchain?tab=readme-ov...

FastAPI example with LangServe (old chat-langchain implementation): https://github.com/langchain-ai/chat-langchain/blob/langserv...

Or just use regular fastapi endpoints instead of langserve.add_routes with a fastapi server: https://fastapi.tiangolo.com/

On: Launch HN: Trellis (YC W24) – AI-powered workflows...

[efriis]

Would recommend using the updated guide here! That link is from the v0.1 docs. https://python.langchain.com/v0.2/docs/how_to/structured_out...

OOC which openai model were you using? Would recommend trying 4o as well as Anthropic claude 3.5 sonnet if ya haven't played around with those yet

[rahimnathwani → efriis]

Thanks.

I was using gpt-3.5-turbo-0125. It was before the recent pricing change.

But I have a bunch of updates to make to the json schema, so will re-run everything with gpt-4o-mini.

Sonnet seems a lot more expensive, but I'll 'upgrade' if the schema changes don't get sufficiently good results.

[efriis → rahimnathwani]

Nice. Could also give haiku a try!

On: LangChain Is a Black Box

[efriis]

Howdy! Erick from LangChain here. Would love to learn a bit more about your bug hunting process and see how we can improve the debugging process for everyone else! Could you email me at erick@langchain.dev?

[siamese_puff → efriis]

Erick, please just read this entire thread.

On: NIST researchers use cellphone compass to measure ...

[efriis]

Kept searching for an “April fools” with a 4/1 publication date to no avail

[YetAnotherNick → efriis]

My first thought as well. But unless they have went too overboard and published a nature paper[1] just for fun, it seems legit.

[1]: https://www.nature.com/articles/s41467-024-47073-2