prvnsmpth

8 ms·

1. 1.▲
   Show HN: Omni – Open-source workplace agent, built on Postgres

   2 points by prvnsmpth 1d ago|0 comments
2. 2.▲
   by prvnsmpth 5mo ago

   Use kimi 2.6 for planning and a cheap model (preferably local) for execution, and then kimi once again for reviewing it. Then finally I review the code. Saves a lot on tokens.
3. 3.▲
   by prvnsmpth 5mo ago

   You can sign up for a plan on the kimi code platform and use it via the pi.dev coding agent, or opencode. In planning, I’d say it’s almost on par with Claude Opus.
4. 4.▲
   by prvnsmpth 7mo ago

   Thanks Philippe! You guys have been super helpful on slack!
5. 5.▲
   by prvnsmpth 7mo ago

   So it depends on the app - e.g., Google has domain-wide delegation where the workspace admin can provide service account creds that allow us to impersonate all users in the workspace and index all their files/email. During indexing, we
6. 6.▲
   by prvnsmpth 7mo ago

   Fair point, although I think we have OpenAI to blame for that - for buying chat.com and pointing it to the most popular textual AI interface of them all :)
7. 7.▲
   by prvnsmpth 7mo ago

   So far both projects are quite similar… the only major difference being the search index. Onyx uses vespa.ai for BM25 and vector search, I decided to go down the Postgres-only route.
8. 8.▲
   by prvnsmpth 7mo ago

   That's a good point, it might make sense to clarify that for individuals who want to self-host. I'll make the change, thanks!
9. 9.▲
   by prvnsmpth 7mo ago

   Exactly, enterprise customers almost always use private model endpoints on their cloud provider for any serious deployments. Data stays within the customer's VPC, data security and privacy is guaranteed by the cloud providers.
10. 10.▲
    by prvnsmpth 7mo ago

    Thank you! Typical RAG implementations I’ve seen take the user query and directly run it against the full-text search and embedding indexes. This produces sub-par results because the query embedding doesn’t really capture fully what the use
11. 11.▲
    by prvnsmpth 7mo ago

    Not yet, there’s a Microsoft connector implementation, but it only does Sharepoint, OneDrive, Outlook etc. and I haven’t tested it thoroughly yet. Teams required some special setup to work IIRC, so I skipped it. Will keep it on the roadmap
12. 12.▲
    by prvnsmpth 7mo ago

    Thanks for sharing! Big part of the reason why I decided on postgres, everything I've read about people using it in prod tells me that most organizations never really grow beyond requiring anything more than what it offers.
13. 13.▲
    by prvnsmpth 7mo ago

    Yeah, Omni uses Postgres and pgvector for search. ParadeDB is essentially just Postgres with the pgsearch extension that brings in Tantivy, a full-text search engine (like Apache Lucene).
14. 14.▲
    by prvnsmpth 7mo ago

    Oops, sorry! That page is still a WIP, haven't pushed it yet. The plan was to expose the main search and chat APIs so that users can build integrations with third-party messaging apps (e.g. Slack), but haven't gotten around to pro
15. 15.▲
    by prvnsmpth 7mo ago

    I've done small scale experiments with up to 100-500k rows, and did not notice any significant degradation in search query latency - p95 still well under 1s. I haven't directly compared against Elasticsearch yet, but I plan to do
16. 16.▲
    by prvnsmpth 7mo ago

    Thank you! Currently permissions are handled in the app layer - it's simply a WHERE clause filter that restricts access to only those records that the user has read permissions for in the source. But I plan to upgrade this to use RLS i
17. 17.▲
    Show HN: Omni – Open-source workplace search and chat, built on Postgres(github.com)

    177 points by prvnsmpth 7mo ago|42 comments
18. 18.▲
    by prvnsmpth 1y ago

    https://casepro.club A platform for consulting aspirants to practice business case interviews. Finding case prep partners is a major pain point for B-school students/consulting aspirants. Fortunately, frontier AI models are
19. 19.▲
    Show HN: Fine-tuning an LLM on your code for better code completions(prvn.sh)

    4 points by prvnsmpth 2y ago|0 comments
20. 20.▲
    The Animated Transformer: An Intuitive Explanation of the Transformer Model(prvnsmpth.github.io)

    3 points by prvnsmpth 3y ago|1 comments
21. 21.▲
    by prvnsmpth 3y ago

    I wrote a short article explaining the Transformer model and how it works, using the Manim library for generating animations. Please read through if the topic interests you, and leave your thoughts and feedback!
22. 22.▲
    by prvnsmpth 4y ago

    It doesn’t make sense to have your build tool call another build tool to actually perform the build. In that case, why I would not just use Gradle directly?
23. 23.▲
    by prvnsmpth 4y ago

    Congrats on building and launching your project! That said, let me just go ahead and share the obvious: https://xkcd.com/936/
24. 24.▲
    by prvnsmpth 5y ago

    The botanical classification doesn’t matter - coconut oil is primarily saturated fat which is what makes it better for you over the PUFA that comes from seed oils.
25. 25.▲
    by prvnsmpth 5y ago

    There should be an 'X' at the top right, is it missing for you?
26. 26.▲
    by prvnsmpth 5y ago

    Thanks for pointing it out! Gotta fix this on Android...
27. 27.▲
    by prvnsmpth 5y ago

    Yes! I plan to add that next.
28. 28.▲
    by prvnsmpth 5y ago

    Thanks :) I'm going to update the dictionary to use the same one as Wordle's, so that should be fixed.
29. 29.▲
    by prvnsmpth 5y ago

    Yeah I guess so! It isn't too much work :)
30. 30.▲
    by prvnsmpth 5y ago

    Agreed, tracking time is an interesting idea!