Hacker News: imaurer

New comment by imaurer in "Delve – Fake Compliance as a Service"

imaurer — Fri, 20 Mar 2026 18:54:45 +0000

vibe compliance

I call them "Loop Bots"

imaurer — Tue, 15 Jul 2025 22:04:04 +0000

Article URL: https://www.imaurer.com/blog/posts/2025-07-15-i-call-them-loop-bots/

Comments URL: https://news.ycombinator.com/item?id=44576314

Points: 5

# Comments: 2

New comment by imaurer in "Prelude – a tiny CLI tool building context prompts from your code"

imaurer — Sun, 21 Jul 2024 13:11:57 +0000

Have a bunch of Makerile commands (pbcopy-api, pbcopy-ui, pbcopy-curr) that use some mishmash of git ls-files, grep, xargs tail -n +1 piped into pbcopy.

Kitchen sink command: pbcopy-all: git ls-files | xargs tail -n +1 | pbcopy

Works like a charm in Q2 2024.

I’m sure this will be a very solved problem by 2025.

New comment by imaurer in "The one about the web developer job market"

imaurer — Wed, 27 Mar 2024 20:35:25 +0000

“ Finding effective documentation, information, and training is likely to get harder, especially in specialised topics where LLMs are even less effective than normal.”

Who needs documentation with Claude and pbcopy?

New comment by imaurer in "Tech Debt: My Rust Library Is Now a CDO"

imaurer — Tue, 26 Mar 2024 14:07:05 +0000

I'm excited for the Michael Lewis version of the Rust library ecosystem.

New comment by imaurer in "OpenAI GPT-4 vs. Groq Mistral-8x7B"

imaurer — Fri, 22 Mar 2024 17:12:49 +0000

Groq will soon support function calling. At that point, you would want to describe your data specification and use function calling to do extraction. Tools such as Pydantic and Instructor are good starting points.

I am collecting these approaches and tools here: https://github.com/imaurer/awesome-llm-json

New comment by imaurer in "Inversion: Fast, Reliable Structured LLMs"

imaurer — Tue, 19 Mar 2024 12:57:27 +0000

Currently, LLM models are not state of the art at Named Entity Recognition. They are slower, more expensive and less accurate than a fine tuned BERT model.

However, they are way easier to get started with using in context learning. Soon, they will be cheaper and probably faster enough too that training your own model will be a waste of time for 95% of use cases (probably higher because it will unlock use cases that wouldn’t break even with the old NLP approaches from a value perspective).

This is why I am tracking LLM structured outputs here:

https://github.com/imaurer/awesome-llm-json

And created an autocorrecting pydantic library that could be used for Named entity linking:

https://github.com/genomoncology/FuzzTypes

FuzzTypes: Pydantic Library for Auto-Correcting Annotation Types

imaurer — Fri, 15 Mar 2024 13:20:15 +0000

Article URL: https://github.com/genomoncology/FuzzTypes

Comments URL: https://news.ycombinator.com/item?id=39715157

Points: 2

# Comments: 0

New comment by imaurer in "Show HN: We built the fastest spreadsheet"

imaurer — Thu, 29 Feb 2024 18:32:51 +0000

R2 support for egress $$ reasons?

New comment by imaurer in "IAC sold 17 apps to Bending Spoons. $100M deal, all 330 employees fired"

imaurer — Fri, 12 Jan 2024 19:50:29 +0000

How'd they get both Chainsmokers?

What Is a GPT?

imaurer — Mon, 08 Jan 2024 22:18:47 +0000

Article URL: https://www.imaurer.com/what-are-gpts/

Comments URL: https://news.ycombinator.com/item?id=38919059

Points: 1

# Comments: 0

New comment by imaurer in "LuaX: A Lua Dialect with JSX"

imaurer — Sun, 31 Dec 2023 14:10:37 +0000

I don’t know anything about Lua other than I want to try it out because of redbean [1]. Wonder if this project can work with that?

[1] https://redbean.dev/

New comment by imaurer in "Weaviate – Open-Source AI Native Vector Database"

imaurer — Mon, 29 May 2023 22:26:47 +0000

One feature I haven’t seen people write about is the ref2vec capability. I find this to be an interesting way to get some knowledge graph-like capabilities out of Weaviate.

Posting here to see if someone sees it by happenstance and writes an awesome article about it someday so I can read it.

https://weaviate.io/blog/ref2vec-centroid

New comment by imaurer in "Donut: OCR-Free Document Understanding Transformer"

imaurer — Mon, 29 May 2023 18:25:18 +0000

Two places I use it: Preview on my Mac, photos on my phone. Haven’t seen an api yet.

The Answer Is the Easy Part

imaurer — Thu, 18 May 2023 14:37:52 +0000

Article URL: https://www.imaurer.com/the-answer-is-the-easy-part/

Comments URL: https://news.ycombinator.com/item?id=35988607

Points: 1

# Comments: 0

New comment by imaurer in "Which vector similarity metric should I use?"

imaurer — Tue, 16 May 2023 16:20:45 +0000

Yes

New comment by imaurer in "Which vector similarity metric should I use?"

imaurer — Tue, 16 May 2023 16:05:53 +0000

Yes, cosine distance works best in convex or normalized sets. Thinking about adding this caveat. Thanks for the question.

Which vector similarity metric should I use?

imaurer — Tue, 16 May 2023 15:34:15 +0000

Article URL: https://www.imaurer.com/which-vector-similarity-metric-should-i-use/

Comments URL: https://news.ycombinator.com/item?id=35963261

Points: 2

# Comments: 4

Medical AI: How to raise and eventually make money

imaurer — Mon, 15 May 2023 13:29:44 +0000

Article URL: https://www.imaurer.com/funding-reimbursement-medical-ai/

Comments URL: https://news.ycombinator.com/item?id=35947825

Points: 1

# Comments: 0

New comment by imaurer in "What is a Vector Database? (2021)"

imaurer — Fri, 05 May 2023 12:57:37 +0000

Well Weaviate is graphql and it has filtering and hybrid search which is a great feature that pg can’t fully support because it doesn’t have bm25

https://weaviate.io/developers/weaviate/api/graphql/filters

https://weaviate.io/blog/hybrid-search-explained

I have a ChatGPT session where I have asked it to do a hybrid search using filtering, pg fts and vector search. Looks reasonable just need to test it and write it up somewhere.