I built a bot that plays Pokémon FireRed. It can explore, battle, and respond to game events. Farthest I made it was Viridian Forest.I paused development a couple months ago, but given the launch of ClaudePlaysPokemon, decided to open source!

Show HN: LLM plays Pokémon (open sourced)

Hey everyone! I’m working on Telescope - an open-source web-based log viewer designed to make working with logs stored in ClickHouse easier and more intuitive.I wasn’t happy with existing log viewers - most of them force a specific log format, are tied to ingestion pipelines, or are just a small part of a larger platform. Others didn’t display logs the way I wanted.So I decided to build my own lightweight, flexible log viewer - one that actually fits my needs.Check it out:<pre><code> Video demo: https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=5IItMOXwugY

 GitHub: https:&#x2F;&#x2F;github.com&#x2F;iamtelescope&#x2F;telescope

 Live demo: https:&#x2F;&#x2F;telescope.humanuser.net

 Discord: https:&#x2F;&#x2F;discord.gg&#x2F;rXpjDnEc</code></pre>

Show HN: Telescope – an open-source web-based log viewer for logs in ClickHouse

Libredesk is a 100% free and open-source customer support desk, the backend is written in Go and the frontend is in Vue JS with ShadnCN for UI components.Unlike many &quot;open-core&quot; alternatives that lock essential features behind enterprise plans, Libredesk is fully open-source and plans to always stay this way.It&#x27;s in alpha (v0.1.0) right now, but there’s a working demo available. I built this because I wanted a truly open and self-hosted alternative to platforms like Chatwoot, Intercom, and Zendesk.Would love feedback, suggestions, and thoughts from the community.GitHub: <a href="https:&#x2F;&#x2F;github.com&#x2F;abhinavxd&#x2F;libredesk">https:&#x2F;&#x2F;github.com&#x2F;abhinavxd&#x2F;libredesk</a>Demo: <a href="https:&#x2F;&#x2F;demo.libredesk.io&#x2F;" rel="nofollow">https:&#x2F;&#x2F;demo.libredesk.io&#x2F;</a>

Show HN: Libredesk – Open-source customer support desk. Single binary app

It took me a few years of fits and starts but I&#x27;ve finally got my DIY persistence of vision bike light to display video!I designed and built the PCB for it and wrote the code for the wifi streaming of video frames with a bit of help from Claude. This was actually one thing I was dreading as I was hitting a wall being burned out on this project. With a little bit of Claude&#x27;s help in getting the ESP32 wifi stack working It&#x27;s finally finished!Some tech specs:
Using an ESP32 
42 LED&#x27;s per strip, 4 strips total spaced 90 deg apart. 
Angular resolution of the image is 1 deg so it updates all 168 LED&#x27;s , 360 times per rotation.The LED&#x27;s are SK9822 individually addressable LED&#x27;s (also known as DotStar from Adafruit).Uses a magnet and hall effect sensor to keep track of rotation speed.The server can be run from a phone and using termux with a python server to serve locally to the esp.I need to periodically spin it up every time it slows down.My video camera shutter speed needs to be slow to show the correct effect otherwise it is either incomplete or it shows too many &quot;frames&quot; in a video frame.Gonna get around to making a blog post at some point to talk more about it in detail!

Show HN: Wireless video streaming on POV bike display

Recently created a minimal persistent relational database in Go. Main focus was on implementing &amp; understanding working the of database, storage management &amp; transaction handling. Use of B+ Tree for storage engine(support for indexing), managing a Free List (for reusing nodes), Support for transactions, Concurrent Reads.
Still have many things to add &amp; fix like query processing being one of the main &amp; fixing some bugsRepo link - <a href="https:&#x2F;&#x2F;github.com&#x2F;Sahilb315&#x2F;AtomixDB">https:&#x2F;&#x2F;github.com&#x2F;Sahilb315&#x2F;AtomixDB</a>Would love to hear your thoughts

Show HN: A Database Written in Golang

Hi HN ! This is an open source project I&#x27;ve been working on for a while. It&#x27;s actually the third breakout-like game I&#x27;ve built. I really like the simplicity of the concept, having just one input (the puck horizontal position) and a simple gameplay that gives a low skill floor.My girlfriend and I played a lot of the excellent LBreakoutHD, an open source breakout game that follows the traditional formula of having multiple lives, scoring points by breaking bricks and catching the good upgrades that spawn randomly and fall down, while avoiding bad upgrades.She liked this game because it is non-violent, and doesn&#x27;t make her sick like the first person 3D games. There are some issues though, it gets boring to break the last bricks, it&#x27;s a bit unfair or slow sometimes, and the run length is too long with 30 something levels to clear for a high score.I wanted to make a clone that would be fix those issues. I first tried to introduce more strategy by making the upgrades visible from the start, instead of them appearing randomly. You&#x27;d then strategize what to break first to earn more points. That first version is playable (<a href="https:&#x2F;&#x2F;breakout-v1.lecaro.me&#x2F;" rel="nofollow">https:&#x2F;&#x2F;breakout-v1.lecaro.me&#x2F;</a>) but a bit too complex.I then wanted to simplify the gameplay, but make the game multiplayer in split screen (<a href="https:&#x2F;&#x2F;breakout-v2.lecaro.me&#x2F;" rel="nofollow">https:&#x2F;&#x2F;breakout-v2.lecaro.me&#x2F;</a>). Instead of bonus and malus, each brick drops some coins, and you need to catch them with the puck. This worked pretty well. You can play with the keyboard (A&#x2F;D and LEFT&#x2F;RIGHT keys) or mouse or both. The bomb explosions will blow coins around, including the coins of other players, and if you lose your ball, then a gap opens between your screen and the guy next to you, to give him a chance to lend you his ball. You can play using your phone as a controller by scanning the QR code, but make sure everyone is on the same Wi-Fi and that the the firewalls are down.For my last version (<a href="https:&#x2F;&#x2F;breakout.lecaro.me&#x2F;" rel="nofollow">https:&#x2F;&#x2F;breakout.lecaro.me&#x2F;</a>) is focused on the game feel and juice. It is about breaking bricks and catching coins, like the v2, but you can now pick upgrades at the end of each level. Your score unlocks more upgrades and levels that are added to the pool for the next runs. There are currently 31 upgrades and 91 levels to unlock.Please have a try and tell me what you think. The game should run well on Firefox, safari and chrome on mobile and pc. It is available in F-Droid and on the play store, The source code is on GitLab. All links are in the game menu.

Show HN: Breakout with a roguelite/vampire survivor twist

Show HN: Python micro event loop library (~250 LOC)

i wanted to change the habit of reaching for my phone in the morning and doomscrolling away an hour so i built an app to help me. now i have to literally touch grass before accessing my most distracting appsthe app is built in swiftui, uses the screen time apis provided by apple and google vision to recognise grass or noti&#x27;d love to get your thoughts on the concept.

Show HN: I built an app to stop me doomscrolling by touching grass

Wampy is an interface addon for modern Linux-based Walkmans, which allows you to switch between standard interface and custom one using hardware Hold switch.The project was born out of handful of standard UI nitpicks and &quot;can I make a prettier UI?&quot;. There is no Rockbox port for my device (NW-A55), so I did a UI myself, unlocking and adding various features along the way, such as:<pre><code> - Winamp 2 skin support
 - Custom cassette skins
 - Digital clock skin ported from iPod Nano 7g
 - Audio amplification table editor for S-Master HX
 - All audio filters are available regardless of firmware
 - Per-song audio filter options
 - Standard interface enhancements (add clock and increase cover art size)
 - Low latency USB DAC module
 - FM radio on devices with FM chip and Walkman One (A40&#x2F;50)
</code></pre>
The result covers 6 models, from cheapest NW-A30 to premium NW-WM1Z. Development process involved a lot of reverse engineering, digging into device internals and was pretty fun overall; there are links to development stories in README.md, describing how this or that feature was added.

Show HN: Wampy, interface addon for Linux-based Walkmans

Hi HN, I’m Phil, one of the co-founders building emdash. Previously, I was an early engineer at Facebook and led Customer Products at Square.We’ve focused on making chat and video work together so distributed teams can stay aligned without drowning in information. You can try it here: <a href="https:&#x2F;&#x2F;emdash.io" rel="nofollow">https:&#x2F;&#x2F;emdash.io</a>.It frustrated us how easily important conversations would happen and then disappear. Slack never quite matched how we worked. Channels were too coarse which led to noisy notifications and broken search. Zoom meetings weren’t much better–unless someone took perfect notes (which rarely happened), video calls became black holes of lost knowledge.We spent too much time trying to find the information we needed to do our jobs.To address this, we’re testing a few concepts and would appreciate your feedback on the value of the following:(1) Automatically record, summarize, and transcribe your team’s video chats. We store meeting content directly inside discussions to facilitate search and discovery.(2) Make it easy to manage &amp; organize conversations of varying scope. A chat between team members can be forked into a dedicated Discussion with its own audience permissions and subscription. Individual messages or entire Discussions can be moved after the fact. Conversations can evolve unpredictably, so having the right tools to keep them organized post-hoc was important to us.(3) Improve search with AI and hierarchical information retrieval. We use LLMs to uncover insights, summarize content, and connect the dots across related discussions, meetings, and documents. You can ask questions like “What are the team’s priorities this week?” or “What did we decide to do with feature X?” and get back a generative response AND deep links into the original chats and meetings.Try it out: <a href="https:&#x2F;&#x2F;emdash.io" rel="nofollow">https:&#x2F;&#x2F;emdash.io</a> and tell us what you think!

Show HN: Emdash – Slack/Zoom alternative for distributed team collaboration

Show HN: MyCoder, an open source Claude-Code alternative

Hello HN!
Keeping Elasticsearch or Algolia in sync with Postgres is a pain when your data spans multiple tables. Traditional CDC tools like Debezium capture table-level events but don’t help when you need to index a full entity composed of related records. That&#x27;s why I built PG-Capture.PG-Capture lets you define a schema as complex as you need, aggregate table-level events, and reconstruct structured objects spanning multiple tables. It then publishes them as meaningful domain-level events to any queue (RabbitMQ, SQS, etc.).PG-Capture does NOT replace Debezium, it works on top of it (or any other event source). This keeps it flexible and tool-agnostic.I&#x27;d love to hear your thoughts! Does this solve a problem you&#x27;ve run into? Would you use something like this in your stack? Feedback is welcome!

Show HN: PG-Capture – a better way to sync Postgres with Algolia (or Elastic)

As someone living outside a country with a popular currency, I hated wanting a product but not being able to buy it due to a different currency. Either it&#x27;s not allowed or there are immense fees.So I built World-Price, an embeddable pricing table generator that integrates with stripe to generate clean prices (either .99 prices or .00 prices) for every currency to be able to accept every user.Right now I have released a somewhat MVP, and would love your feedback if this provides use to you. Enjoy!

Show HN: World-Price – Making international pricing transparent and reassuring

Since I was young, I&#x27;ve loved anime, and over the years, manga has brought me joy, given me courage, and sparked excitement in my heart.However, as I read more, I realized that many of these manga weren&#x27;t translated at all. I also came across some AI-based translation tools, but the results often fell short.So, I decided to create a tool that allows manga fans to read and enjoy their favorite manga, no matter the language or whether a translation team is available.This product has just been launched, and there are certainly areas that can be improved. However, with time, I&#x27;m confident it will only get better.You&#x27;re welcome to try it out and share your valuable feedback!

Show HN: Instantly Translate Manga – TranslateManga

Backend and API testing can be tedious—especially when dealing with multi-step workflows, deep validation of API responses, and complex edge cases. We&#x27;ve built qapir.io to make this easier, offering a no-code solution for defining and running backend and API tests without writing any code.What It Offers:* No-Code Test Creation – Define backend tests with an intuitive YAML-based syntax.* Support for Complex Scenarios – Easily test multi-step API workflows and chained API calls.* Human-Readable Reporting – Clear, structured test reports that make debugging easier.* Support for multiple protocols - Currently supports plain HTTP and GraphQL, planning to support: SQL, Redis, Kafka, etc.* An actively developing set of features – Upcoming features: Dashboard with Test-Results, Configurable HTTP-Mocks, and a solution for receiving and validating webhooks - Webhook-InterceptorWhy Use It:If you need a way to test backend services without maintaining a ton of code, qapir.io provides a simple, structured approach. It’s designed to handle everything from basic API checks to more advanced multi-step scenarios—all while keeping tests readable and easy to maintain.Download it for free at <a href="https:&#x2F;&#x2F;qapir.io" rel="nofollow">https:&#x2F;&#x2F;qapir.io</a>, and start writing your tests in minutes!

Show HN: Simplifying Backend Testing with qapir.io

Hey HN,We&#x27;ve been experimenting with a real-time, version-controlled NoDB for Deno &amp; React called GoatDB. The idea is to remove backend complexity while keeping apps fast, offline-resilient, and easy to self-host.Runs on the client – No backend required, incremental queries keep things efficient.
 Self-hosted &amp; lightweight – Deploy a single executable, no server stack needed.
 Offline-first &amp; resilient – Clients work independently &amp; can restore state after server outages.
 Edge-native &amp; fast – Real-time sync happens locally with minimal overhead.Why We Built It:
We needed something that’s simpler than Firebase, lighter than SQLite, and easier to self-host. GoatDB is great for realtime collaboration, offline, prototyping, single-tenant apps, or ultra-low-cost multi-tenant setups—without backend hassles.Would love feedback from HN:* Are there specific features or improvements that would make it more useful?* How do you handle similar problems today, and what’s missing in existing solutions?If you&#x27;re interested in experimenting or contributing, the repo is here:
 GitHub Repo: <a href="https:&#x2F;&#x2F;github.com&#x2F;goatplatform&#x2F;todo">https:&#x2F;&#x2F;github.com&#x2F;goatplatform&#x2F;todo</a>Looking forward to your thoughts!

Show HN: GoatDB – A lightweight, offline-first, realtime NoDB for Deno and React

We’ve been working on a search problem that requires querying both text and numbers simultaneously. For example, in a dataset of clothing items with descriptions and prices, a search for “slim pants for $20” should prioritize skinny jeans for $25 over slim pants for $50 because they are semantically similar and the price is closer. I’ve found that standard embedding models struggle with numerical ordering, while text-to-SQL methods rely on exact matches and often filter out too many results.To solve this, we built a system designed specifically for structured datasets like CSVs or tables. Here’s a demo link where you can upload a small CSV to try out (no login required): <a href="https:&#x2F;&#x2F;demo.tryvoker.com" rel="nofollow">https:&#x2F;&#x2F;demo.tryvoker.com</a>.Unlike most RAG approaches, we process each column independently, handling text with embeddings and numbers with custom scoring. When a user submits a query, we parse it into relevant fields—for instance, extracting “slim pants” as the description and “20” as the price. We then compute cosine similarity between the description embeddings and “slim pants” while also calculating the percent error between the user’s price input and the numerical field. These individual similarity scores are then combined across all columns to generate a final ranking.Right now, our system works best with well-structured data, so some preprocessing is often needed. We’re working on improving this by detecting and restructuring messy data automatically, such as pivoting columns or extracting attributes from large text fields. We’re also adding feedback mechanisms, like a thumbs up&#x2F;down system, to refine future search results based on user input.
I’d love to hear about your experiences with similar search challenges and would appreciate any feedback!

Show HN: Ranked Search for Semi-Structured Data

Hi HN! My team at Efficient Computer (<a href="https:&#x2F;&#x2F;efficient.computer" rel="nofollow">https:&#x2F;&#x2F;efficient.computer</a>) and I are working on a new computer architecture and I am leading the team that is building the compiler for our chip. Our hardware is focused on energy-constrained embedded applications and is super efficient, which lets devices run for years on a small (ie AA) battery. We built a playground for our compiler that has a cool visualizer and debugger that shows how your C code (more languages to come) maps to our Fabric architecture, and an energy model that shows you how much less energy your code uses on our architecture. Check it out: <a href="https:&#x2F;&#x2F;www.efficient.computer&#x2F;resources&#x2F;effcc-compiler-playground-launch" rel="nofollow">https:&#x2F;&#x2F;www.efficient.computer&#x2F;resources&#x2F;effcc-compiler-play...</a>

Show HN: Compiler Playground for energy-efficient embedded dataflow processor

Show HN: LLM Token Visualizer – How big is 128k token input

Show HN: Open-source tool that send UI feedback with context

Show HN: Schedual

Hi HN,I’m a university student interested in Gen AI, and over the holidays, I built a project in public: Who Rates the Rater? – a crowdsourced dataset for benchmarking AI-generated storytelling.The Idea 
Inspired by Chatbot Arena, this lets users compare AI-generated stories and provide feedback. The goal is to refine LLMs for creative writing using real human preferences.How It Works- Live Demo: <a href="https:&#x2F;&#x2F;storycrowdsourcepreference.streamlit.app" rel="nofollow">https:&#x2F;&#x2F;storycrowdsourcepreference.streamlit.app</a>- Tech Stack: Built with Streamlit + Supabase- Open Source: <a href="https:&#x2F;&#x2F;github.com&#x2F;clchinkc&#x2F;story_crowdsource_preference">https:&#x2F;&#x2F;github.com&#x2F;clchinkc&#x2F;story_crowdsource_preference</a>Get Involved- Try it &amp; star the repo if you find it interesting- Bug reports &amp; feature requests welcome on Twitter- Follow me for future AI &amp; storytelling projectsWould love to hear your thoughts!

Show HN: Built a "Story Chatbot Arena" to Crowdsource AI Story Preferences

I often need simple text tools (e.g. diffing, epoch time conversion, number base conversion) and was frustrated that I couldn&#x27;t find a simple option that was as fast as I wanted. Many (SEO-optimized) websites exist, but they unnecessarily send your input to their servers. I used tools like DevToys but their diff has been broken on Mac for a while [1] and it requires installing an app, and CyberChef is good but requires too many interactions (yes, I&#x27;m picky).So, I decided to build my own. txtl tries to autodetect what mode you want based on the input, and in general is fast and gets out of your way as much as possible. For example, if you paste in a number then it will show base conversions, but if it looks like an epoch timestamp then it will display the corresponding date. And of course it&#x27;s open source at <a href="https:&#x2F;&#x2F;github.com&#x2F;viveksjain&#x2F;txtl&#x2F;">https:&#x2F;&#x2F;github.com&#x2F;viveksjain&#x2F;txtl&#x2F;</a>This was partly also a way for me to try using LLMs to build a frontend&#x2F;new project. This could probably be a whole blog post, but I see why some people can say that LLMs can 10x their coding productivity (I certainly didn&#x27;t find this true in my day job) - I was really impressed with what o1 can do, it probably got me 70% of the way there in one shot. And as someone not familiar with frontend development, iterating was way faster. But, it still requires plenty of feedback and requires handholding - I had to completely rewrite the diff rendering logic, no matter what I tried I couldn&#x27;t get an LLM to get it right.Would love feedback on the autodetection and overall usability![1] <a href="https:&#x2F;&#x2F;github.com&#x2F;DevToys-app&#x2F;DevToys&#x2F;issues&#x2F;1296">https:&#x2F;&#x2F;github.com&#x2F;DevToys-app&#x2F;DevToys&#x2F;issues&#x2F;1296</a>

Show HN: Txtl – Fast static website of text utilities

Hi hn - I co-own a diner where I co-host a puzzle night that is kind of like a diner-themed escape room. At the last one, I made a puzzle that was crossword-like clues nested in brackets. People at the diner seemed to like it, so I resolved to make it a real game and Bracket City was born: <a href="https:&#x2F;&#x2F;bracket.city" rel="nofollow">https:&#x2F;&#x2F;bracket.city</a>.I love crosswords, so it&#x27;s been fun to write crossword-like clues:<pre><code> [it contains MSG]
</code></pre>
as well as clues that would not make it into a crossword:<pre><code> [___ &lt;=== you ===&gt; hard place]
</code></pre>
I write all the puzzles and post a new one at midnight ET every day of the week.Still working on a lot of features&#x2F;fixes. I&#x27;m aware that scoring based on keystrokes is pretty unfair, especially given not-ideal custom keyboard on mobile! Still thinking through the best solution there.Also fun fact: if you sign up for the email list, you get a special &quot;Word of the Day&quot; email written by James Somers (of <a href="https:&#x2F;&#x2F;jsomers.net" rel="nofollow">https:&#x2F;&#x2F;jsomers.net</a>). The only way to sign up for the email list is to finish a puzzle!**(answer key: NYC, ROCK)

Show HN: Bracket City – A daily, exploded (?) crossword puzzle

I figured I needed to work on my coding skills before building the next groundbreaking AI app, so I started working on this free tool site. Its basically just an aggregation of various commonly used calculators and unit convertors.Link: <a href="https:&#x2F;&#x2F;www.calcverse.live" rel="nofollow">https:&#x2F;&#x2F;www.calcverse.live</a>Tech Stack: Next, React, Typescript, shadcn UI, Tailwind CSSWould greatly appreciate your feedback on the UI&#x2F;UX and accessibilty. I struggled the most with navigation. I&#x27;ve added a search box, a sidebar, breadcrumbs and pages with grids of cards leading to the respective calculator or unit convertor, but not sure if this is good enough.

Show HN: While the world builds AI Agents, I'm just building calculators

Whilst ChatGPT Deep Research is very useful for generating in-depth reports, it&#x27;s time consuming to copy, reformat the text (thousands of words) and clean referenced hyperlinks for use in a professional context. Out of frustration, I built deep research docs to help save time by automating the reformatting, cleaning links, footnote references, and conversion to shareable PDF format.Hopefully this helps you save time to focus on meaningful work.Let me know your feedback.

Show HN: I built an app to convert ChatGPT Deep Research to PDFs with footnotes

Show HN: Mobile, LLM-Assisted IDE PoC

There have been a few times I wanted the ability to select some text out of a Markdown doc. For example, a GitHub CI check to ensure that PRs &#x2F; issues &#x2F; etc are properly formatted.This can be done to some extent with regex, but those expressions are brittle and hard to read or edit later. mdq uses a familiar pipe syntax to navigate the Markdown in a structured way.It&#x27;s in 0.x because I don&#x27;t want to fully commit to the syntax being stable, in case real-world testing shows that the syntax needs tweaking. But I think the project is in a pretty good spot overall, and would be interested in feedback!

Show HN: Jq-Like Tool for Markdown

Hey HN,I built SendFast (<a href="https:&#x2F;&#x2F;sendfast.app" rel="nofollow">https:&#x2F;&#x2F;sendfast.app</a>) because I was tired of clunky urls, and not knowing when my files are viewed. Google Drive links are long and annoying, WeTransfer has ads and expirations, and most alternatives don’t tell you if someone actually opened your file.With SendFast:- You upload and share files up to 32GB
- Get notifications when someone views or downloads your file (no more “Did you get my email?”)
- Recipients don’t need an account &amp; can preview files (images&#x2F;videos&#x2F;pdf) instantly without ads
- Links won’t expire as long as you need them [also working on better pricing and guarantees to make sure this extends down to even free-tier]It’s built for people who send important files (contracts, pitch decks, legal docs, photos, etc.) and want a simple, reliable alternative that doesn’t lock them into a Big Tech ecosystem.I’m still improving it, so I’d love any feedback. What do you currently use for sharing large files? Any pain points?Check it out if you&#x27;re interested!