▲ 321 points
Ask HN: What are you working on? (August 2026)
What are you working on? What have you been curious about lately?
back
1188 comments
I got laid off at the end of April and have been building my own personal dream tool since, it's a skeuomorphic carpentry simulator with an agent MCP. You start with real wood specs and a shop full of miter saws, router tables, table saws and the like. Think of it like Sketchup but no extruding rectangles.
It's been really fun to build with. Agents can create parametric procedures so e.g. stick framing becomes a simple function call for the next agent, building a flywheel. There is full agent-human parity in every surface, so there is a human UI for the procedures too.
Every operation is yaml, so agents pick it up pretty quickly. Agents can also file feature requests when they reach for a tool that doesn't exist. You can export to 3d printing files, video walk-through, step files, get a BOM and a cut plan, view your project in life-size with AR. There is a notion-like interface for authoring build guides... a lot going on. If you're a woodworker/software engineer you might enjoy it.
You can add the MCP to your agent of choice, send a picture, description, sketch or whatever; say "build this with Sawdust"
Love it, I had a similar but probably worse idea a few years back while woodworking. I agree the issue with essentially all CAD and sketchup etc is it forces you to work with primitives that are different from your real world ones. Extruding shapes as you said, instead of actually milling lumber and working in common known dimensions.
I also suspect there’s a fairly finite number of sane join primitives, you can have rabbets, dados, dominos, dovetails, pocket screws, etc. These could be experimented with digitally against an existing design. And your tool availability could determine what you can do.
And of course cut order and assembly is just a DAG and can be generated from geometric constraints of the actual design.
Wishing you all the best with this project. Super cool.
If you are looking to improve the mcp, you should try adding a UI to it (see https://apps.extensions.modelcontextprotocol.io/api/). And the server happens to be in python, you should check out: https://mplemay.github.io/belgie/examples/mcp/. It is a project I have been working on which makes it easy to build MCP UIs from python.
Have been building something non-CAD too in the space which is also aligned with carpentry. This looks really neat and can't believe how similar the functionality is. Would be great to chat!
Cool project. A couple of years ago I was gifted "One-Plank Woodworking Projects" by Andy Standing for Christmas (which is great). It's a book of plans for things you can build with, obviously, one standard plank of wood. That might be a decent addition to the filters: rather than project type or tools available, what wood does the user has to hand.
Hey, looks like we both have a thing with 3D, love it!
Mine is about bouldering, check it out https://news.ycombinator.com/item?id=49234599
Very cool! I was just today thinking I should have an AI help me design a new piece of furniture I've been thinking about, and send it off to a CNC shop.
The MCP server seems to have a bug. The tool list flashes for a second in Claude Desktop, but then immediately states "This connector has no tools available." and the error "Couldn’t reload tools from the server." when I try to refresh it.
This is amazing I been wanting something like for a long time. I'm also curious whats up with the no VC logo at the bottom of the page?
Awesome! Built a sliding keyboard desk with a drawer for an 88 key MIDI controller. I've been wanting to custom build one of these for a while, this will help!
https://app.sawdust.diy/share/b5362e61-39bc-4fa5-a537-38ca78...
This sounds a lot like a project I’ve started a few times and never made it far with. I called it woodlang. It was the idea of CAD but rather than designing the shape and figuring how to make it that shape, you specify a series of shop operations like blender geometry nodes.
This was all pre AI. Super impressive that you’ve made something awesome. I’ll try it out tonight
Lots of fun projects being listed. I'm working on Preloop(https://preloop.dev), a way to run your unmodified Github Actions locally or self-hosted in isolated cross-platform microvms(uses smolvm which uses libkrun vmm).I got frustrated with Github Actions reliability and inability to run or test my workflows locally so i re-wrote the runner and reverse-engineered the control plane so it follows the exact runner protocol the offficial runner does. Each job runs in a microvm that boots in 400ms from a packed image. Also supports a way to pause on every failure so you or your agent can immediately shell at that point, fix and retry(in a new environment) to verify. Also supports the full DAP protocol so you can interactively step/forward/investigate context at each step/job.
Try it out: https://github.com/preloopdev/preloop
My two biggest projects are:
https://www.idlequest.net/ - a from-scratch recreation of the 1999 MMORPG classic EverQuest, featuring a real-time MMO server written in Go, a 3D game browser-based client built with Babylon.js, and the clasic UI recreated using React. You can play it both as a regular MMORPG but also in "idle mode" where it basically plays itself. This had up to 70 players online at once a while back which was really cool, though its popularity is in a bit of a lull right now. There are still tons of bugs and stuff that isn't fully implemented (and balance issues) but it's still fun for most people who have nostalgia for it.
https://www.newyokosuka.com/ - an unofficial Shenmue MMO experience. Based on the 1999 Sega Dreamcast game Shenmue, it's playable entirely in your browser! It's basic world exploration with forklift shenanigans, some basic dialogue systems, and a basic combat system test, but I'm about to push an update that includes all cutscenes and story progression as well. The story part is taking a bit of time because I'm having to mold it to fit in an MMO context. But it's a fun project to work on and the Shenmue community loves it so far.
I've been building the most advanced automotive diagnostic app on the market
I used to be a mechanic, and spend a lot of my free time working on project vehicles, motorcycles, tractors, trucks etc. I wanted a tool to track the tasks I needed done, then a way to easily access VIN info for parts search, then I DMV records, and finally built an iphone app to scan the OBD2 diagnostic codes and hook them up to a custom ai diagnostic prompt.
It's been a blast to work on, I need to focus on marketing it and get some users that aren't close friends, but it's been serving my purposes this summer so far.
I've seen nothing with anywhere close to this depth of diagnostic reporting.
https://crewchief.cc https://apps.apple.com/us/app/crewchief-auto/id6760673109
A boat game for 5+ age group!
I was never much of a game developer, but recently my 5 year old son wanted to play a game so we decided to build one together.
It's a boat game, you basically steer a boat in an isometric 2d world, transporting passengers, cargo etc between ports.
As we designed it, the ideas just kept on coming. He made or directed a lot of the art inside of the game (even the audio), and we included friends and family as characters within the game.
A lot of fun to build - hilarious at times. I used Lua and LÖVE2D, building basically a game engine for an isometric world, so that took significantly more time than expected.
If you want to test it [1], please do, but keep in mind it's only in Norwegian, but I am sure most people would be able to figure it out as it's meant for age group 5 and up:
[1] https://apps.apple.com/no/app/b%C3%A5tspillet/id6791630382
At this point our entire extended family and their kids, grandchildren etc are actually playing this, so it's been super fun.
I have been working on and off on https://www.whichtemple.in
- it is a collection of ~ 2700 temples across India, with images sourced from wikipedia
- it shows you a random temple each time you visit
- it has a map view of temples across India
So this is just the tip of the iceberg. I am trying to build a catalogue of all temples, starting with India, and expanding on to the rest of the world. The focus is on temples originating from Indic religions like Hinduism, Jainism and Buddhism.
I have personally discovered many cool temples that I have never heard of, but I am struggling on how to gamify this. Feedback welcome!
The final stretch of getting my game ready for release next week: https://store.steampowered.com/app/2764460/Sandustry/
It's an automation game and factory builder with "falling sand" physics. Think Noita meets Factorio.
I’m working on a native UI framework in rust. I want to be able to build cross platform, fully native applications with native code. And I want platform native components.
I started with Leptos - which gives you a nice declarative UI syntax and reactive elements. I used Claude to get that working on iOS, cocoa and GTK. I made a nice little iOS schedule app for attendees at conferences. It works great. All rust, native UI and a fast and small binary. The current UI code is here: https://github.com/josephg/leptos-native (not officially part of leptos). It works well, but not all of the components I want have been implemented.
At the moment I’m rewriting the core to ditch the vibe coded parts and break the whole thing into an app backend API and separate platform layer. I want to be able to write apps in wasm and have a windows / mac / Linux host that can run them using native UI. And a runner to run any app using a TUI. And have a compilation process to compile the whole thing into a single native process for app developers and to publish to the App Store.
I'm working on Dictatorship Simulator. A deckbuilder where, instead of directly attacking your enemy, you manipulate a political simulation to do it for you. You start out on the fringe, and slowly work your way up, fighting an increasingly worried and panicky government - as well as your own.
https://store.steampowered.com/app/4564800/Dictatorship_Simu...
I'm building an uptime monitoring service, in Elixir, called https://larm.dev. It's a tough field because for some reason people vibe coding apps think this is very easy to do, and there are new apps coming out pretty much every day. I'm still figuring out how to effectively stand out in the noise.
I really enjoy working on the app though. The backend is Elixir, with a Phoenix LiveView dashboard. I've got a cluster of 3 nodes running, talking to each other. The probes that check the monitoring targets are tiny go binaries, spread across hosting providers across the world. I've also got synthetic probes now, running in little playwright sandboxes.
A lot of the challenges have been around how to distribute work across the probes, with different strategies for the basic TCP/HTTP probes vs the synthetic workers, and how to effectively store large amounts of events (Clickhouse).
Still plugging away on Cassava, a CSV editor that sits somewhere between sublime text and Excel: https://cassava.dev/. Our most recent releases added first-class xls/xlsx file opening and an autofill handle.
The project goals are simple:
* Treat CSV files like first-class spreadsheets; don't do anything that isn't supported by serializing to/from CSV files
* Be as performant as your favorite text editor while retaining the ergonomics of a tabular data editor
I hope you check us out.
My wife and I are working on a math/science/CS-inspired jewelry business. We try to create pieces that stand on their own aesthetically but have a hidden meaning.
We currently have two styles: lambda calculus based pieces (we depict the lambda/Tromp diagram) where we have Y-Combinator earrings (well, strictly speaking they are one beta reduction away from Y-combinator. Aesthetic oblige) and a pendant depicting a lambda expression computing Graham's number. The other style is quantum computing circuits, based on quantum computing research my brother (a physics professor) is doing: a pendant that is actually a non-local controlled-NOT gate. I wrote a tiny DSL to describe the jewelry pieces, and an interpreter to produce CAD files. We then either 3D print them or have them produced by lost-wax.
We are 200% out of our comfort zone (and love it): I know nothing of front end dev, payments, or anything like that. The diamond district in New York is a neighborhood we normally actively avoid, but if you are forced to go there it is fascinating (people examining diamonds on the corner of the street, others in fur coats in summer straight out of a mafia movie...), and especial marketing. Jewelry is a completely saturated business (luckily we are not doing this to pay the rent); we think we have a unique angle, but we are still figuring out the target audience (if there is one), how to advertise, etc.
Any ideas on marketing are welcome.
Store: https://studio-galois.com/
I'm working on https://wireplug.org:
wireplug lets you create mesh VPNs using standard WireGuard. Your wg config files stay almost the same, except you no longer have to specify an `Endpoint` for your peers. Instead, wireplug detects all your possible endpoints (including LAN) and coordinates them with your peers for you. It handles NAT traversal and works with the in-kernel WireGuard driver on Linux and OpenBSD. This lets you keep managing your own keys and network topology, while also maintaining connectivity when moving between networks.
Currently working on improving NAT traversal by adding support for PCP and NAT-PMP.
Working on open sourcing my data analytics canvas: https://kavla.dev/
I've worked on it for about a year, with 1700 commits. I've been going back and forth if I should just open up the canvas or also the multiplayer experience.
I've always had the dream to make some money from a side project, but I don't think this one is it, so I'll just make the entire repo public. Just want to clean up some stuff first.
Lots of fun tech used: duckDB WASM, Cloudflare Durable Objects, tldraw and pocketbase
I'm going whole hog on LLM-driven development to build out an old school FOSS forum that can run in resource-constrained environments with a minimum of fuss. Think phpBB but written in Rust with modern UI niceties and a SQLite backend. My hope is that interested parties will be able to deploy this on random SBCs or Unikraft and use cloudflared to expose their service to interested parties. Watching old forums that I used to frequent get gobbled up by VerticalScope makes me pine for the days of small self-hosted communities that existed for the hell of it instead of monetized interaction.
I created a make-your-own-circuit simulator in python to help learn electronics. The best part about this is that it is entirely hosted online and people can just go through the lessons, learn electronics by skipping the maths, and in the process have made their own circuit simulator (which is arguably a pretty cool project on its own) at the end of it.
I’m exploring how to use AI for faith oriented experiences and products. There are two specific areas I’m interested in. I’m furthest along with this app in the App Store. https://trysojourn.app
1) grounding LLM scripture references in the actual text and eliminating hallucinations. This is the platform for that. https://withlattice.com
2) how to use AI to get people into scripture more as opposed to being a substitute for it. This is the product I made for it - it uses #1 above. https://trysojourn.app
I've been working on a Persian classical poetry archive with LLM-generated translation. As a poetry enthusiast but a non-speaker of Persian, I wanted to read Rumi, Hafiz, and Bedil, but found myself jumping between ChatGPT and Google Translate. I sourced my data from Ganjoor, which was luckily available via their public API. Then I quickly put together a website with bring-your-own-key translation feature.
For now, there's no backend and it's hosted on GitHub Pages; it's a static website reading chunks from an sqlite-based DB file. I intend to change it in the future by running an LLM (preferably a frontier Gemini model, as works great for Persian) over the whole data set so you won't need a key to read the translated poetry. However, I don't have the budget of $1000-$2000 to spend on the tokens currently.
Website: https://0x5ce.github.io/farsidaan
Code: https://github.com/0x5CE/farsidaan
P.S. Contributions welcome.
I've been building https://consciousnesslibrary.org, an open library for psychedelic and consciousness research that pulls from over a dozen academic APIs every 20 minutes, dedupes them, organizes by relevance, and synthesizes evidence.
I just finished my master's in Psychedelics and Consciousness Studies, and built the library I wanted while studying. I started it over a year ago, but put it down until my practicum advisor suggested I dust it off and go hard on it during my last term.
As of this writing, there are over 35,700 papers by over 82,200 authors organized into 37 topics, growing daily. There's a 2D map of the whole corpus and evidence syntheses on the topic pages, plus the ability to generate your own syntheses with any prompt.
It's built with Rails 8.1, uses one Postgres (for full-text search, vectors, and job queue), and runs on a $24/mo DigitalOcean droplet that I just upgraded from the $12 tier and deploy to with Kamal. Total LLM spend so far is ~$28 in DeepSeek-V4-Flash, trending down after a big one-time backfill push.
Relevance was the hardest part to get right by far. Ensuring we keep papers on LSD the psychedelic vs Lumpy Skin Disease, or Ketamine for depression vs anesthesia in cats was no small feat. The design ended up being a cheap keyword prefilter, then an LLM rubric that accepts or rejects papers. More on that here: https://consciousnesslibrary.org/docs/article-pipeline.html
It's free, no ads, no signup to read anything, and a registered 501(c)(3) I fund myself for now. I'd really like to know what breaks or any other feedback, and would love to answer any questions about it!
Rated Analysis, rating relative chess analysis: https://rated-analysis.pickle-dev.com/
I've been a bit obsessed with chess lately and have read/heard things like "X is the top engine move, but no one under master plays that line". To me this begs the questions "which line are players below master playing" and "which line are players at/above/below my rating playing". This question motivated me to build a tool to analyze the move distribution (win/draw/loss rate) from a given position across rating buckets (400, 500, 600... 3000). Luckily, https://database.lichess.org/ provides an open/free database of games that have just the data required to build such a tool.
The Rated Analysis tool consists of an 1) an offline job to compute/store the positional move distribution bucketed by rating, and 2) the https://rated-analysis.pickle-dev.com/ webpage to serve the data along with some nice to haves like loading user games from platforms like chess.com/lichess.org.
Climbing gym photogrammetry. My idea is that route and climb recordings is best recorded against a 3D climbing gym model, rather than a plan video, and technology has gotten good enough to do with a regular phone.
As a climber myself, to climb effectively, one has to review their own climb, and with other people. 4D video, body positions across time, is very helpful in improving climbing. I managed to get a reasonable result (far from good) from one video, rather requiring multiple camera angles.
Haven't gotten to monetization yet, but I am enjoying every moment of making it.
A read only preview at https://kmcheung12.github.io/climb-preview/w/ae43c6e5
I have been working on calculating and predicting lottery numbers and am currently publishing the results on Zenodo; however, my work is in Brazilian Portuguese—comprising 19 academic technical articles—which you can check out here: https://zenodo.org/search?q=metadata.creators.person_or_org....
The last time somebody made this thread, I posted about https://the-waterline.com/, a site I launched as a side project/eval to see how well Claude can do a decent-sized project (mostly) independently. It pulls public government data about water levels in the western US, updates the website and sends out a newsletter.
It started getting some search impressions and clicks, so I decided to launch the next 9 of the top 10 ideas that ChatGPT came up with for this type of site - won't post all of them here, but https://the-dwell.com/ (data on ship and truck dwell times/port congestion) is doing the best thus far from a traffic perspective.
First I had it send me a weekly analysis of the GA4/GSC data to see what could be improved about the site, but I've since automated that - each week it reviews traffic and search queries, then updates articles or creates new ones to meet demand. So now in theory I have a full engine to launch these sites and have them self-improve over time. Kinda neat to see all of that being done autonomously.
The thing that these sites really need are some backlinks, so I'm thinking through how to get those. I had ChatGPT generate some lists of places that might find real value in linking to these sorts of things, but I'm unwilling to let an AI agent start emailing people. I'm going to try sending some emails to some of the places that seem like a really good fit, see what the response is, and if it's good try to figure out how to do that on a larger scale without AI email spam.
I'm working on DB Pro [1]
A modern database client with dashboards, saved queries, workflows, and yes, AI built into it.
I've focused on building brilliant UX over the last year, and it has paid off. I've had two emails saying that they have cancelled their Jetbrains subscription for DataGrip which has been so validating.
It's been a labour of love and I'm super happy to have disrupted even a small part of the ginormous market that is database clients.
I left my job as a principal software engineer a few months back and have been working on an agentic harness and https://blackbear.app -- a local first, end-to-end encrypted everything app with file editors for notes, documents, sheets, free-hand canvases, slides, and a whole lot more. Also features a calendar and task system, and a lightweight discovery engine. There's a BYOK agent system (Pax) that works with you across the whole app (OpenAI, Anthropic, Gemini, and local Ollama support). Works in a browser, from the CLI, and from native mac and linux apps (mobile on the way). Not exactly sure my end goal yet, but it's been fascinating building these things
I’ve been working on my comprehensible input reader for Spanish still. I launched the beta exactly last month and it has been going well.
The project combines the idea of graded readers with spaced repetition systems like Anki. You generate stories with LLMs that are tailored to your Spanish level, then click the words you don’t know. As you read more stories, the system brings those words back naturally in future stories so you keep seeing them in context.
That’s how it works on a high level but behinds the scenes, there is a good amount of natural language processing going on, majority with the spaCy package. I’m hoping one day to post a blog about it on here.
It’s been awesome to see it actually being used by people and I have had fun talking with users and implementing feedback. At my current job, I usually don’t get that level of customer interaction so it’s been nice.
I’m hoping to take it out of beta soon and launch a paid tier with higher story generation limits, while keeping a free tier available. The goal is mainly to start covering the cost of running the project and keep improving it.
If you’re learning Spanish and have a A2 level or higher feel free to check it out. And even if you don’t know Spanish feel free to ask me questions about it, I would love to talk more detail about it.
I finished wiring up the backend data merge so user(s) (aka me, really) can record trees as they walk and have those merged into the general city data on my urban tree map app[1]. There's some very cool trees in the neighborhood I'm excited to have recorded.
Which leads me to trying to get the processing off github actions (it's all duckdb pipelines), which has led me to improve the orchestration implementation of my 'better sql' [2] to optimize the asset based refresh flow. I'm now wiring that up in a general cloud service to give me the data orchestration platform I've always wanted. Building a multi-tenant cloud experience has been interesting!
[1]: https://greenmtnboy.github.io/sf_tree_reporting [2]: https://github.com/trilogy-data/pytrilogy
I'm working on Eland (Extensible Layered Algorithm for Network Drawing), a graph layout library strongly inspired by Elk, but a complete reimplementation.
The reason I need a complete reimplementation is that existing graph layout libraries tend to be black boxes that you can't really fiddle with. If you want to tweak part of the algorithm, you have to fork it or write your own. With Eland, you can replace individual steps with your own implementation while keeping the rest of the algorithm unchanged. To ensure validity of the various steps, I've introduced contracts where every step specifies what properties it requires from the graph, which properties it promises to add, and which it invalidates, which I think is pretty cool.
The reason I need all of this is that I'm working on a system to display and browse through data lineage graphs. I've noticed through experience that existing graph layout libraries don't do quite what I need, so I ended up building my own.
I used to teach C to students many years ago, and with agents it's becoming a niche craft. C is still both an important and fun language to use, and I've been very slowly making a C language course to let beginners learn a bit about the language and practice writing simple code with no AI: https://c-course.com
A work in progress, and will stay so for much longer, I guess.
Afrika Labs (https://afrikalabs.org) — open speech-to-text for Kikuyu (Gĩkũyũ, ~8M speakers, my mother tongue), as a start, then other African languages.
Almost none of Africa's 2,000+ languages have working ASR. I'm fine-tuning Whisper on ~217h of transcribed Kikuyu (no Kikuyu token exists, so I'm repurposing the Swahili one). A recent paper hit ~26% WER with similar data, so a usable v0 is realistic.
I put my flat in the UK on the market recently and it has given me a lot of insights into how broken the real estate market is!
One thing I found particularly annoying is how dominant rightmove is and how much of the traffic for house hunting they control.
As a small way of standing up to them I created https://rigove.co.uk/ (rightmove without htm - the hidden toll machine ;)
If you change rightmove to rigove while browsing a property for sale you get a page which shows where else that property is listed.
It is only a few days old so it is currently more miss than hit but I plan to reach out to estate agents to increase coverage as quick as I can.
I’m currently on the beach at Google (since mid July), so I’m looking internally and externally for a new role. But I’m exhausted, so I’m playing around with local LLMs and DeepSeek more as well. I’m currently trying to get an agent harness going where orchestration happens via bazel, and agents are completely isolated beyond what context you give them access to as expressed in starlark. Agent executions are organized as a DAG like a build would be, where the output of one agent can serve as context for another (there is also a feedback mechanism which is needed for arbiter nodes).
I’ve mentioned it before, but we can get some pretty good results for developing implementation and tests with this setup. The context isolation makes it easy to separate test and code writing using a specification as a source of truth. Tests are then run by an arbiter that feedback to tests or implementation depending on who it blames for test failures. The tests are then treated as an independent implementation, and through the math of coincident errors, the arbiter feedback loop ensures that both are eventually correct (N-versions and clean room research did this with humans in the 80s). This then allows us to use less reliable LLMs to develop code, like Qwen on a MacBook Pro.
I hope to have something out on GitHub before my beach period ends (well, if I’m unsuccessful in getting a new role, then I’ll have a lot more time to play at least).
AI quant trading (so that I have another income and can stop being threatened by managers that if I don't meet their unrealistic expecations that I won't have food on the table next year).
So far it's going well, 30000% gains on my initial capital in 3 months very gradually over 100K trades. But I need to build some more sleeves because (a) the edge could die any time (b) I hit capacity of that edge and it can only grow linearly now.
Working on this [0]
Trying to make the world’s smallest maze solving neural network. Why? Because it’s fun. Why aren’t you?
Current champion model is 14 bytes - it solves 96.5% of unseen mazes upto 21x21.
Try it yourself on the website. Also a work in progress write up at [1]
[0] - https://minimio.ai
I built and run several SaaS platforms:
- https://dave-bot.com -> a full-stack AI platform where you can generate videos, images, music, code, 3d objects with frontier Gen AI models.
- https://headsnap.io -> a platform that you can generate images of yourself based on 4 selfies.
- https://quantiq.live -> a service providing financial and historical data for stocks, as well as government trades.
- https://aivestor.tech -> an AI agent that picks small/midcap stocks and trades them using Alpaca API. It uses Reddit, news, polymarket, Google Trends and many other data sources to take investment decisions.
- @Polyglot_lingua_bot -> a voice-enabled Telegram-based bot that can help you learn new languages.
- https://select.supply -> a directory of carefully-curated and well-crafted products.
All of those allowed me to quit my day job and live a comfortable and flexible life. I still invest time in maintenance and adding new features, but I love coding, marketing and everything that comes with promoting and selling a SaaS (and I also have a serious addiction for Stripe notifications).
On top of that, I developed my own software agency where I help clients build and scale software (https://bitheap.ch).
https://www.play-when.com/daily
When? Is a timeline game I've been working on for a bit. It's a board game I love playing with my family and I wanted a slicker mobile version I could play with them.
A new spec, TPX, which is meant to enable building SaaS where the user brings their own inference:
It has a few potentially interesting side effects:
- enterprises can plug their preferred LLM into the SaaS they use.
- self hosters can extend their rigs into the web.
- simple SaaS products can offer LLM integrations without their entire pricing model being upended by LLM hosting and infrastructure.
- we can all live like Ender in Xenocide and have a personal AI that has context on everything we are doing.
- have a no op provider for people who are anti llm.
I'm working on Trebella [0], a tool to help classical music students structure their practice sessions more effectively. I tried to build something similar ~10 years ago in the form of a more interactive music game, but felt like it wasn't actually that helpful for intermediate and advanced students. So this time I'm doing it in collaboration with actual music teachers and building a tool that actually helps me learn. I'd say 90% of my flute practice is done using it at this point, and I'm in the process of taking on the first arms length users in a closed beta. Please shoot me an email if you are interested in any way!
I'm working on what could be described as a Ralph Loop on horse steroids.
It's an externally orchestrated (meaning it's not the agent running the loop) spec driven agentic build system.
A bunch of buzz words but essentially it's a tool you use as the last step after planning your work and writing a spec, to actually have agents write the code.
But it happens within an orchestrator that is durable and which coordinates layers of agents to build -> verify -> review -> triage -> fix -> verify fixes , until the code you get as a result is as close as possible to exactly what you specified.
So you can plan however you want, use whatever harness or agent etc and then just delegate the build to the engine (MCP server or CLI)
You can mix and match models from any provider for any part of it.
I'm building this because I think micromanaging agents is what causes AI burnout and because I want to be able to plan work, walk away and come back knowing it's been correctly implemented.
https://engine.build if you want to be notified for the launch, it's going to be free to use and fully OSS (Apache 2.0)
I’m making Stacks, an open source ebook manager and server, as a lightweight alternative to Calibre.
Written in Swift, it consists of a SwiftUI browser and a Mac/Linux server and CLI. Shares are automatically discovered via Bonjour/avahi and also published as OPDS feeds.
It can import Calibre libraries and has basic support for managing books on Kindles, but I’ve only tested it with a MTP Paperwhite as that’s all I’ve got.