Show HN: Sourcebot – Self-hosted Perplexity for your codebase
14 by bshzzle | 1 comments on Hacker News.
Hi HN, We’re Brendan and Michael, the creators of Sourcebot ( https://ift.tt/aACFRoB ), a self-hosted code understanding tool for large codebases. We originally launched on HN 9 months ago with code search ( https://ift.tt/rvWYhQF ), and we’re excited to share our newest feature: Ask Sourcebot. Ask Sourcebot is an agentic search tool that lets you ask complex questions about your entire codebase in natural language, and returns a structured response with inline citations back to your code. Some types of questions you might ask: - “How does authentication work in this codebase? What library is being used? What providers can a user log in with?” ( https://ift.tt/G4BgALK ) - “When should I use channels vs. mutexes in go? Find real usages of both and include them in your answer” ( https://ift.tt/qIgYBu0 ) - “How are shards laid out in memory in the Zoekt code search engine?” ( https://ift.tt/N2K3gkM ) - "How do I call C from Rust?" ( https://ift.tt/MrYvzHy ) You can try it yourself here on our demo site ( https://ift.tt/tDUsZ0F ) or checkout our demo video ( https://youtu.be/olc2lyUeB-Q ). How is this any different from existing tools like Cursor or Claude code? - Sourcebot solely focuses on code understanding . We believe that, more than ever, the main bottleneck development teams face is not writing code, it’s acquiring the necessary context to make quality changes that are cohesive within the wider codebase. This is true regardless if the author is a human or an LLM. - As opposed to being in your IDE or terminal, Sourcebot is a web app. This allows us to play to the strengths of the web: rich UX and ubiquitous access. We put a ton of work into taking the best parts of IDEs (code navigation, file explorer, syntax highlighting) and packaging them with a custom UX (rich Markdown rendering, inline citations, @ mentions) that is easily shareable between team members. - Sourcebot can maintain an up-to date index of thousands of repos hosted on GitHub, GitLab, Bitbucket, Gerrit, and other hosts. This allows you to ask questions about repositories without checking them out locally. This is especially helpful when ramping up on unfamiliar parts of the codebase or working with systems that are typically spread across multiple repositories, e.g., micro services. - You can BYOK (Bring Your Own API Key) to any supported reasoning model. We currently support 11 different model providers (like Amazon Bedrock and Google Vertex), and plan to add more. - Sourcebot is self-hosted, fair source, and free to use. Under the hood, we expose our existing regular expression search, code navigation, and file reading APIs to a LLM as tool calls. We instruct the LLM via a system prompt to gather the necessary context via these tools to sufficiently answer the users question, and then to provide a concise, structured response. This includes inline citations, which are just structured data that the LLM can embed into it’s response and can then be identified on the client and rendered appropriately. We built this on some amazing libraries like the Vercel AI SDK v5, CodeMirror, react-markdown, and Slate.js, among others. This architecture is intentionally simple. We decided not to introduce any additional techniques like vector embeddings, multi-agent graphs, etc. since we wanted to push the limits of what we could do with what we had on hand. We plan on revisiting our approach as we get user feedback on what works (and what doesn’t). We are really excited about pushing the envelope of code understanding. Give it a try: https://ift.tt/AmIv0zN . Cheers!
Thursday, July 31, 2025
Wednesday, July 30, 2025
Tuesday, July 29, 2025
Monday, July 28, 2025
Sunday, July 27, 2025
New top story on Hacker News: Ask HN: What are you working on? (July 2025)
Ask HN: What are you working on? (July 2025)
46 by david927 | 111 comments on Hacker News.
What are you working on? Do you have any new ideas you're thinking about?
46 by david927 | 111 comments on Hacker News.
What are you working on? Do you have any new ideas you're thinking about?
Saturday, July 26, 2025
Friday, July 25, 2025
Thursday, July 24, 2025
New top story on Hacker News: Show HN: A code editor that integrates into the browser
Show HN: A code editor that integrates into the browser
6 by quintu5 | 0 comments on Hacker News.
When the startup I was working for shut down, I knew it would probably be a while before my health allowed me to commit to a new role, so I decided to start working on some personal projects to keep my mind active and engaged. With AI-augmented VS Code forks being all the rage at the time, I wanted to take a slightly different angle on a code editor/viewer using the same core technology. That led me to building Tachi Code, a Monaco-based code editor that integrates directly into your browser as an extension to streamline your more ephemeral coding tasks, so you can spend less time switching between your code editor and browser. The original flow that piqued my interest was viewing raw source files or API responses. Historically, I've used a JSON formatter extension to prettify JSON, but I wanted something more powerful, more universal, and quite frankly, something that looked better, so I built Tachi Code with the ability to detect when you're viewing pre-formatted text and inject itself into the page, so it's always beautifully syntax highlighted, foldable, and regex searchable. Then I added context menu integrations, so you could quickly edit snippets, compare text, or view the current page's source in Tachi Code's editor. The browser extension works offline with the only external HTTP requests going to GitHub to retrieve JSON Schemas or additional themes. All user data stays local. The only tracking is CloudFlare's web analytics beacon on TachiCode.dev (not present in the browser extension or in the EU). TachiCode.dev is a sandbox environment that serves the latest commit of Tachi Code's editor hosted on CloudFlare Pages. The core stack is: - React 19 - Monaco Editor - Radix UI - Zustand - Shiki - WXT (full SBOM is available via the about dialog if you want to dig deeper) Monaco Editor provides the code and diff editors, as well as low level systems for configuration and theming. There's a lot of hackery involved in surfacing those systems and integrating them into the larger React app. Shiki is used to provide more complete syntax highlighting than Monaco Editor provides out of the box. The rest of the UI is primarily based on Radix UI components, typically starting from a shadcn template and then reworked to use colors provided by the theme system. Zustand is my go-to for any kind of shared/persistent state. WXT just turns browser extension development and publishing into a breeze. If you've got any feedback or a question about how the app was developed, I'd love to hear it!
6 by quintu5 | 0 comments on Hacker News.
When the startup I was working for shut down, I knew it would probably be a while before my health allowed me to commit to a new role, so I decided to start working on some personal projects to keep my mind active and engaged. With AI-augmented VS Code forks being all the rage at the time, I wanted to take a slightly different angle on a code editor/viewer using the same core technology. That led me to building Tachi Code, a Monaco-based code editor that integrates directly into your browser as an extension to streamline your more ephemeral coding tasks, so you can spend less time switching between your code editor and browser. The original flow that piqued my interest was viewing raw source files or API responses. Historically, I've used a JSON formatter extension to prettify JSON, but I wanted something more powerful, more universal, and quite frankly, something that looked better, so I built Tachi Code with the ability to detect when you're viewing pre-formatted text and inject itself into the page, so it's always beautifully syntax highlighted, foldable, and regex searchable. Then I added context menu integrations, so you could quickly edit snippets, compare text, or view the current page's source in Tachi Code's editor. The browser extension works offline with the only external HTTP requests going to GitHub to retrieve JSON Schemas or additional themes. All user data stays local. The only tracking is CloudFlare's web analytics beacon on TachiCode.dev (not present in the browser extension or in the EU). TachiCode.dev is a sandbox environment that serves the latest commit of Tachi Code's editor hosted on CloudFlare Pages. The core stack is: - React 19 - Monaco Editor - Radix UI - Zustand - Shiki - WXT (full SBOM is available via the about dialog if you want to dig deeper) Monaco Editor provides the code and diff editors, as well as low level systems for configuration and theming. There's a lot of hackery involved in surfacing those systems and integrating them into the larger React app. Shiki is used to provide more complete syntax highlighting than Monaco Editor provides out of the box. The rest of the UI is primarily based on Radix UI components, typically starting from a shadcn template and then reworked to use colors provided by the theme system. Zustand is my go-to for any kind of shared/persistent state. WXT just turns browser extension development and publishing into a breeze. If you've got any feedback or a question about how the app was developed, I'd love to hear it!
Wednesday, July 23, 2025
Tuesday, July 22, 2025
New top story on Hacker News: Show HN: Any-LLM – lightweight and open-source router to access any LLM Provider
Show HN: Any-LLM – lightweight and open-source router to access any LLM Provider
27 by AMeckes | 11 comments on Hacker News.
We built any-llm because we needed a lightweight router for LLM providers with minimal overhead. Switching between models is just a string change : update "openai/gpt-4" to "anthropic/claude-3" and you're done. It uses official provider SDKs when available, which helps since providers handle their own compatibility updates. No proxy or gateway service needed either, so getting started is pretty straightforward - just pip install and import. Currently supports 20+ providers including OpenAI, Anthropic, Google, Mistral, and AWS Bedrock. Would love to hear what you think!
27 by AMeckes | 11 comments on Hacker News.
We built any-llm because we needed a lightweight router for LLM providers with minimal overhead. Switching between models is just a string change : update "openai/gpt-4" to "anthropic/claude-3" and you're done. It uses official provider SDKs when available, which helps since providers handle their own compatibility updates. No proxy or gateway service needed either, so getting started is pretty straightforward - just pip install and import. Currently supports 20+ providers including OpenAI, Anthropic, Google, Mistral, and AWS Bedrock. Would love to hear what you think!
Monday, July 21, 2025
Sunday, July 20, 2025
Saturday, July 19, 2025
Friday, July 18, 2025
Thursday, July 17, 2025
Wednesday, July 16, 2025
Tuesday, July 15, 2025
Monday, July 14, 2025
Sunday, July 13, 2025
Saturday, July 12, 2025
Friday, July 11, 2025
New top story on Hacker News: Show HN: RULER – Easily apply RL to any agent
Show HN: RULER – Easily apply RL to any agent
2 by kcorbitt | 0 comments on Hacker News.
Hey HN, Kyle here, one of the co-founders of OpenPipe. Reinforcement learning is one of the best techniques for making agents more reliable, and has been widely adopted by frontier labs. However, adoption in the outside community has been slow because it's so hard to implement. One of the biggest challenges when adapting RL to a new task is the need for a task-specific "reward function" (way of measuring success). This is often difficult to define, and requires either high-quality labeled data and/or significant domain expertise to generate. RULER is a drop-in reward function that works across different tasks without any of that complexity. It works by showing N trajectories to an LLM judge and asking it to rank them relative to each other. This sidesteps the calibration issues that plague most LLM-as-judge approaches. Combined with GRPO (which only cares about relative scores within groups), it just works (surprisingly well!). We have a full writeup on the blog, including results on 4 production tasks. On all 4 tasks, small Qwen 2.5 models trained with RULER+GRPO beat the best prompted frontier model, despite being significantly smaller and cheaper to run. Surprisingly, they even beat models trained with hand-crafted reward functions on 3/4 tasks! https://ift.tt/JqXsny3 Repo: https://ift.tt/Ktp08kA
2 by kcorbitt | 0 comments on Hacker News.
Hey HN, Kyle here, one of the co-founders of OpenPipe. Reinforcement learning is one of the best techniques for making agents more reliable, and has been widely adopted by frontier labs. However, adoption in the outside community has been slow because it's so hard to implement. One of the biggest challenges when adapting RL to a new task is the need for a task-specific "reward function" (way of measuring success). This is often difficult to define, and requires either high-quality labeled data and/or significant domain expertise to generate. RULER is a drop-in reward function that works across different tasks without any of that complexity. It works by showing N trajectories to an LLM judge and asking it to rank them relative to each other. This sidesteps the calibration issues that plague most LLM-as-judge approaches. Combined with GRPO (which only cares about relative scores within groups), it just works (surprisingly well!). We have a full writeup on the blog, including results on 4 production tasks. On all 4 tasks, small Qwen 2.5 models trained with RULER+GRPO beat the best prompted frontier model, despite being significantly smaller and cheaper to run. Surprisingly, they even beat models trained with hand-crafted reward functions on 3/4 tasks! https://ift.tt/JqXsny3 Repo: https://ift.tt/Ktp08kA
Thursday, July 10, 2025
Wednesday, July 9, 2025
Tuesday, July 8, 2025
Monday, July 7, 2025
New top story on Hacker News: Show HN: I Got Tired of Calculator Sites, So I Built My Own
Show HN: I Got Tired of Calculator Sites, So I Built My Own
8 by calculatehow | 7 comments on Hacker News.
I’ve always found that online calculators tend to have bad UIs, especially on mobile. Most of the calculator websites I’ve come across use outdated and inconvenient ways of inputting data, or they format the results in confusing ways. I’ve noticed that fraction calculators (especially mixed fractions) are terrible to use, even on desktop. I haven’t built one of those yet, but it’s something I’m planning to tackle soon. This is a project I’ve always wanted to work on, but I’m relatively new to this space. So far, I’ve created a collection of simple calculators focused on math and finance. I’d really appreciate any feedback on the UI/UX or anything else you think could be improved. You can try it here: https://ift.tt/Qxw0Ujy
8 by calculatehow | 7 comments on Hacker News.
I’ve always found that online calculators tend to have bad UIs, especially on mobile. Most of the calculator websites I’ve come across use outdated and inconvenient ways of inputting data, or they format the results in confusing ways. I’ve noticed that fraction calculators (especially mixed fractions) are terrible to use, even on desktop. I haven’t built one of those yet, but it’s something I’m planning to tackle soon. This is a project I’ve always wanted to work on, but I’m relatively new to this space. So far, I’ve created a collection of simple calculators focused on math and finance. I’d really appreciate any feedback on the UI/UX or anything else you think could be improved. You can try it here: https://ift.tt/Qxw0Ujy
Sunday, July 6, 2025
Saturday, July 5, 2025
Friday, July 4, 2025
Thursday, July 3, 2025
Wednesday, July 2, 2025
Tuesday, July 1, 2025
Subscribe to:
Posts (Atom)