The best Hacker News stories from Show from the past day
Latest posts:
Show HN: Piazza, syncing the web in a vectorized DB
We recently released an open source project that syncs wikipedia with a vector database : <a href="https://github.com/Piazza-tech/Piazza-Updater">https://github.com/Piazza-tech/Piazza-Updater</a><p>We used Verba, Weaviate and Docker for deployment<p>We'd like to have some feedback on how to continue the project, which data sources would be interesting to vectorize. You can give feedback on our landing page <a href="http://piazza.tech" rel="nofollow">http://piazza.tech</a> Please leave a star !
Show HN: Real-Time YOLO Object Detection in Elixir: Fast, Simple, Extensible
Show HN: Real-Time YOLO Object Detection in Elixir: Fast, Simple, Extensible
Show HN: Real-Time YOLO Object Detection in Elixir: Fast, Simple, Extensible
Show HN: Prompt Engine – Auto pick LLMs based on your prompts
Nowadays, a common AI tech stack has hundreds of different prompts running across different LLMs.<p>Three key problems:<p>- Choices, picking from 100s of LLMs the best LLM for that 1 prompt is gonna be challenging, you're probably not picking the most optimized LLM for a prompt you wrote.<p>- Scaling/Upgrading, similar to choices but you want to keep consistency of your output even when models depreciate or configurations change.<p>- Prompt management is scary, if something works, you'll never want to touch it but you should be able to without fear of everything breaking.<p>So we launched Prompt Engine which automatically runs your prompts for you on the best LLM every single time with all the tools like internet access. You can also store prompts for reusability and caching which increases performance on every run.<p>How it works?<p>tldr, we built a really small model that is trained on datasets comparing 100s of LLMs that can automatically pick a model based on your prompt.<p>Here's an article explaining the details: <a href="https://jigsawstack.com/blog/jigsawstack-mixture-of-agents-moa-outperform-any-single-llm-and-reduce-cost-with-prompt-engine" rel="nofollow">https://jigsawstack.com/blog/jigsawstack-mixture-of-agents-m...</a>
Show HN: Prompt Engine – Auto pick LLMs based on your prompts
Nowadays, a common AI tech stack has hundreds of different prompts running across different LLMs.<p>Three key problems:<p>- Choices, picking from 100s of LLMs the best LLM for that 1 prompt is gonna be challenging, you're probably not picking the most optimized LLM for a prompt you wrote.<p>- Scaling/Upgrading, similar to choices but you want to keep consistency of your output even when models depreciate or configurations change.<p>- Prompt management is scary, if something works, you'll never want to touch it but you should be able to without fear of everything breaking.<p>So we launched Prompt Engine which automatically runs your prompts for you on the best LLM every single time with all the tools like internet access. You can also store prompts for reusability and caching which increases performance on every run.<p>How it works?<p>tldr, we built a really small model that is trained on datasets comparing 100s of LLMs that can automatically pick a model based on your prompt.<p>Here's an article explaining the details: <a href="https://jigsawstack.com/blog/jigsawstack-mixture-of-agents-moa-outperform-any-single-llm-and-reduce-cost-with-prompt-engine" rel="nofollow">https://jigsawstack.com/blog/jigsawstack-mixture-of-agents-m...</a>
Show HN: App to discover job listings directly from company websites
Show HN: Countless.dev – A website to compare every AI model: LLMs, TTSs, STTs
Show HN: Countless.dev – A website to compare every AI model: LLMs, TTSs, STTs
Show HN: Countless.dev – A website to compare every AI model: LLMs, TTSs, STTs
Show HN: BunkerWeb – The Open-Source Web Application Firewall (WAF)
Show HN: BunkerWeb – The Open-Source Web Application Firewall (WAF)
Show HN: BunkerWeb – The Open-Source Web Application Firewall (WAF)
Show HN: Hacker Herald – like HN but with crowdsourced pics and subtitles
Hi folks this is a project I worked on with some of my students when I was running an online JS programming course. Although the online course is no more, I finally got around to releasing Hacker Herald with a former instructor and student - thanks Arnav and Archis!<p>To those wondering if there is a need for such a Hacker News front end, I would just point out that most newspaper websites are laid out like this - clearly some people like this kind of layout!<p>Also for some stories a picture really does help - currently there is a HN story titled, "Portland airport grows with expansive mass timber roof canopy". But IMO it's better to actually see a picture of the timber roof while scrolling rather than having to click through to the article.
Show HN: Hacker Herald – like HN but with crowdsourced pics and subtitles
Hi folks this is a project I worked on with some of my students when I was running an online JS programming course. Although the online course is no more, I finally got around to releasing Hacker Herald with a former instructor and student - thanks Arnav and Archis!<p>To those wondering if there is a need for such a Hacker News front end, I would just point out that most newspaper websites are laid out like this - clearly some people like this kind of layout!<p>Also for some stories a picture really does help - currently there is a HN story titled, "Portland airport grows with expansive mass timber roof canopy". But IMO it's better to actually see a picture of the timber roof while scrolling rather than having to click through to the article.
Show HN: Hacker Herald – like HN but with crowdsourced pics and subtitles
Hi folks this is a project I worked on with some of my students when I was running an online JS programming course. Although the online course is no more, I finally got around to releasing Hacker Herald with a former instructor and student - thanks Arnav and Archis!<p>To those wondering if there is a need for such a Hacker News front end, I would just point out that most newspaper websites are laid out like this - clearly some people like this kind of layout!<p>Also for some stories a picture really does help - currently there is a HN story titled, "Portland airport grows with expansive mass timber roof canopy". But IMO it's better to actually see a picture of the timber roof while scrolling rather than having to click through to the article.
Show HN: Hacker Herald – like HN but with crowdsourced pics and subtitles
Hi folks this is a project I worked on with some of my students when I was running an online JS programming course. Although the online course is no more, I finally got around to releasing Hacker Herald with a former instructor and student - thanks Arnav and Archis!<p>To those wondering if there is a need for such a Hacker News front end, I would just point out that most newspaper websites are laid out like this - clearly some people like this kind of layout!<p>Also for some stories a picture really does help - currently there is a HN story titled, "Portland airport grows with expansive mass timber roof canopy". But IMO it's better to actually see a picture of the timber roof while scrolling rather than having to click through to the article.
Show HN: Hacker Herald – like HN but with crowdsourced pics and subtitles
Hi folks this is a project I worked on with some of my students when I was running an online JS programming course. Although the online course is no more, I finally got around to releasing Hacker Herald with a former instructor and student - thanks Arnav and Archis!<p>To those wondering if there is a need for such a Hacker News front end, I would just point out that most newspaper websites are laid out like this - clearly some people like this kind of layout!<p>Also for some stories a picture really does help - currently there is a HN story titled, "Portland airport grows with expansive mass timber roof canopy". But IMO it's better to actually see a picture of the timber roof while scrolling rather than having to click through to the article.
Show HN: Vekos – a Rust OS with Built-In Cryptographic Verification
VEKOS(Verified Experimental Kernel OS) is a Rust-based experimental kernel that focuses on runtime verification and security. Every filesystem operation and memory allocation generates cryptographic proofs that can be verified in real-time, ensuring system integrity. Think of it like a blockchain for OS operations rather than just storage verification.<p>Key features:
- Merkle tree-backed filesystem with operation verification
- Memory manager with CoW and proof generation
- Buddy allocator with zone-based memory management
- Shell with VT100 support
- Verified boot sequence with stage attestation<p>The kernel is still in alpha, but it demonstrates a novel approach to OS security by making verification a first-class citizen. All critical operations (memory allocations, filesystem writes, process creation) generate proofs that are stored in an append-only chain, similar to a blockchain but for system operations.<p>GitHub: <a href="https://github.com/JGiraldo29/vekos">https://github.com/JGiraldo29/vekos</a><p>I would be excited to get feedback on this project, especially on the verification approach and potential use cases. If you have any question the innerworkings of the development, just ask and I will gladly answer all questions. The code is under the Apache 2.0 license and welcomes contributors.
Show HN: Replace "hub" by "ingest" in GitHub URLs for a prompt-friendly extract
Gitingest is a open-source micro dev-tool that I made over the last week.<p>It turns any public Github repository into a text
extract that you can give to your favourite LLM easily.<p>Today I added this url trick to make it even easier to use!<p>How I use it myself:
- Quickly generate a README.md boilerplate for a project
- Ask LLMs questions about an undocumented codebase<p>It is still very much work in progress and I plan to add many more options (file size limits, exclude patterns..) and a public API<p>I hope this tool can help you
Your feedback is very valuable to help me prioritize
And contributions are welcome!