HackerTrans
TopNewTrendsCommentsPastAskShowJobs

jellyotsiro

no profile record

Submissions

[untitled]

1 points·by jellyotsiro·22 dni temu·0 comments

Molotov Coctail Thrown at Sam Altman's House

nbcnews.com
5 points·by jellyotsiro·3 miesiące temu·1 comments

Show HN: Agentsearch – browse any docs as a filesystem

agentsearch.sh
5 points·by jellyotsiro·3 miesiące temu·0 comments

Show HN: Crust – A CLI framework for TypeScript and Bun

github.com
95 points·by jellyotsiro·4 miesiące temu·38 comments

Show HN: Nia CLI, an OSS CLI for agents to index, search, and research anything

github.com
1 points·by jellyotsiro·4 miesiące temu·0 comments

[untitled]

1 points·by jellyotsiro·4 miesiące temu·0 comments

[untitled]

1 points·by jellyotsiro·5 miesięcy temu·0 comments

[untitled]

1 points·by jellyotsiro·5 miesięcy temu·0 comments

[untitled]

1 points·by jellyotsiro·5 miesięcy temu·0 comments

Show HN: Open-source semantic search over your local notes via CLI

github.com
9 points·by jellyotsiro·5 miesięcy temu·3 comments

[untitled]

1 points·by jellyotsiro·5 miesięcy temu·0 comments

[untitled]

1 points·by jellyotsiro·6 miesięcy temu·0 comments

[untitled]

1 points·by jellyotsiro·6 miesięcy temu·0 comments

Show HN: AI agent that searches the Cursor forum

cursor.trynia.ai
2 points·by jellyotsiro·6 miesięcy temu·0 comments

Show HN: OSS AI agent that indexes and searches the Epstein files

epstein.trynia.ai
211 points·by jellyotsiro·6 miesięcy temu·97 comments

[untitled]

1 points·by jellyotsiro·7 miesięcy temu·0 comments

Launch HN: Nia (YC S25) – Give better context to coding agents

trynia.ai
131 points·by jellyotsiro·7 miesięcy temu·87 comments

comments

jellyotsiro
·4 miesiące temu·discuss
tens of KBs (v small)
jellyotsiro
·4 miesiące temu·discuss
good point.

we’re using “framework” intentionally because it goes beyond argument parsing. crust handles parsing, but also:

type inference across args + flags end to end compile-time validation (so mistakes fail before runtime) plugin system with lifecycle hooks (help, version, autocomplete, etc.) composable modules (prompts, styling, validation, build tooling) auto-generates agent skills and modules from the CLI definitions

so it sits a layer above a traditional arg parser like yargs or commander, closer to something like oclif, but much lighter and bun-native.
jellyotsiro
·4 miesiące temu·discuss
thanks for flagging! the post itself works, just the link at the bottom
jellyotsiro
·4 miesiące temu·discuss
will fix in the next hour!
jellyotsiro
·4 miesiące temu·discuss
thanks for the catch, what we meant is that we’re not committing to strict stability guarantees yet, so APIs may still change as we iterate toward 1.0.
jellyotsiro
·4 miesiące temu·discuss
one of the examples would be trynia.ai (search and index api for ai agents)

here is github: github.com/nozomio-labs/nia-cli
jellyotsiro
·5 miesięcy temu·discuss
sick
jellyotsiro
·6 miesięcy temu·discuss
rag is not a core! we use both semantic search but combining with fts, grep, direct read, etc.
jellyotsiro
·6 miesięcy temu·discuss
Open source models with minimal safety fine tuning or Grok
jellyotsiro
·6 miesięcy temu·discuss
Thanks for testing this. The Bannon email from June 30, 2019 is in there (HOUSE_OVERSIGHT_029622). Good stress test idea.

Couple things happening:

Semantic search limitation: Less-famous names don't have strong embeddings, so it defaults to general connections rather than specific mentions Keyword search gap: You're right — raw grep can catch exact names I'm missing
jellyotsiro
·6 miesięcy temu·discuss
yes! once for files come out, I will add them right away
jellyotsiro
·6 miesięcy temu·discuss
Shareable conversations would definitely make the tool more useful yeah. I really like the query parameter approach over UUIDs so it would make links human-readable
jellyotsiro
·6 miesięcy temu·discuss
On the limited dataset: Completely agree - the public files are a fraction of what exists and I should have mentioned that it is not all files but all publicly available ones. But that's exactly why making even this subset searchable matters. The bar right now is people manually ctrl+F-ing through PDFs or relying on secondhand claims. This at least lets anyone verify what is public.

On LLMs vs traditional NLP: I hear you, and I've seen similar issues with LLM hallucination on structured data. That's why the architecture here is hybrid:

- Traditional exact regex/grep search for names, dates, identifiers - Vector search for semantic queries - LLM orchestration layer that must cite sources and can't generate answers without grounding
jellyotsiro
·6 miesięcy temu·discuss
sorry all publicly available files *
jellyotsiro
·6 miesięcy temu·discuss
it uses semantic search so yes
jellyotsiro
·6 miesięcy temu·discuss
ill take like 1 hour in the evening to dive deeper, i was never familiar with epstein stuff until i built the agent to simplify things for me.
jellyotsiro
·6 miesięcy temu·discuss
Trump famously told New York Magazine in 2002: "I've known Jeff for 15 years. Terrific guy. He's a lot of fun to be with. It is even said that he likes beautiful women as much as I do, and many of them are on the younger side."

Trump and Epstein were social acquaintances in Palm Beach and New York circles during the 1990s-early 2000s. They socialized together at Mar-a-Lago and other venues
jellyotsiro
·6 miesięcy temu·discuss
oh forgot about it, thanks. just a funny project i build in couple hours so didnt really sweat haha
jellyotsiro
·7 miesięcy temu·discuss
thanks! yes writing docs is pain the ass haha
jellyotsiro
·7 miesięcy temu·discuss
The economics are simple. When an agent guesses, it produces wrong code, failed runs, and wasted time. External context is the biggest source of those mistakes, because IDEs only index what’s in your repo. We are a complement to Cursor, ChatGPT, and Claude, not a replacement.