alt Hacker NewsAsk HN: What are you working on? (November 2026)
What are you working on? What have you been curious about lately?
Comments
I'm working on an iPhone app called PhotoGather. https://apps.apple.com/us/app/photogather/id6738232616
I've got 40k+ photos on my iphone, and wanted a better way to interact with them. The basic idea I had was to virtually stitch all the photos together into a gigapixel image, and navigate that image with pinch and zoom. On import, the app hierarchically groups photos based on date and similarity. The user can go back and edit the groups, change the layout, and adjust the size of photos if they choose.
The interesting I noticed is that in the process of curating and organizing my photos, I stumble across memories and moments that I had completely forgot.
Development Challenge: The app does two interesting things underneath the hood... First, given an arbitrary number of thumbnails with different aspect ratios, it finds the best sized rectangle that completely fits all the photos. This allows me to recursively stitch groups together and always output a rectangular thumbnail for the group. Second, the pinch/zoom UI uses these nested thumbnails directly without slicing/dicing to fixed sized tiles.
Hunting bugs in the Python ecosystem using fuzzers and AI tools I develop. If you have a Python package or extension you'd like analyzed, just let me know.
https://github.com/ReviewToolkits
ColliePWA[1] a mobile interface to continue working with your agents when you're on the. Trying to meet devs where they work on the desktop in multiplexers (herdr/tmux/zellij). Remote access can be set up with any front door like tailscale.
Still not sure where this is going as the space is super crowded and moving fast, but it's solving a lot of my own problems and doesn't rely on third parties to run.
APIKulture. An alternative to Postman/Insomnia/Bruno.
To be honest, I started this because Postman and Insomnia became forced-corporate-cloud not usable for me anymore, and Bruno lacked a feature I considered fundamental for my workflow.
The feature is to remember the JSONPath expressions between requests, a feature that Bruno has implemented already.
But I created this thing and I'm using it daily since then. C++ and wxWidgets, it uses 40MB of RAM, and it is super fast, instead of the 400MB used by the above mentioned programs.
This month I'm working on several projects:
1. Broadcast (https://sendbroadcast.net) - Self hosted email marketing software. Been working on this since 2024 and positive revenue. Releasing v3 this week! :)
2. Render Screenshot (https://renderscreenshot.com) - Screenshot API (very much a scratch my own itch project)
3. Widgetful (https://widgetful.com) - A menubar app for managing multiple Claude Code agent sessions
https://www.abejora.com - time tracking for freelancers and small teams
Lately we've been working on access for AI agents. A lot of people want their own agent to do the repetitive part of time tracking. Control remains important, which is why we rebuilt our core this month to support change history and full undo/redo functionality. Feel free to try it with your own agent.
We built this after the timesheet tool we paid for raised its prices and moved some features that we used regularly behind more expensive plans. We had to add three new colleagues, and everything combined this would cost us $500 a year more for the same software.
We run Elixir and Phoenix LiveView on a Kubernetes cluster made of Talos instances.
We're still looking for feedback on the product, so please let us know what you think!
3 products:
- https://polygonjs.com/: a design tool to create interactive 3D configurators
- https://polyreprint.com/: a hosting platform for those configurators. See more enterprise-type examples on https://polygonjs.com/configurators
- https://polyreplay.com/: game made with said design tool
As a self hoster whos paranoid about backups. I wanted to cheat a bit and use s3. With growing resentment on big tech, who I worked for earlier, I'm now building tools that empower myself and other users and decrease reliance of big tech. Arkel is an object store I've been building and testing for the past few months. https://arkeldata.com
1-year part-time experiment: can you build a large project with AI writing the code and AI plus deterministic tools verifying/guiding/enforcing it for structural and behavior consistency, without a human reading the code?
I'm building a polyglot web framework in 15 languages, with C# and Lean4 as the reference implementations. Structural and behavioral audits using deterministic tools check that every port stays consistent with the reference. It's about 1.8M lines now. I have not reviewed the code/neither have i verified the behavior tests except at the time of giving requirement and never seen some of the language code at all.
The question is how far can we produce "working code → correct code [Possible] → good code [Unknown]" with deterministic audits correcting structure, ai mistakes and behavior validation.
So far the quality looks acceptable with lots of correction of AI output done without looking, but it isn't fully verified.
Line counts per language, with changes since August: https://claude.ai/artifact/5f1m1pTM7tZ4boQzwsoAG7
https://github.com/Polign/recall
Polign Recall, a state and memory management for agents that can provide append only state logging and governance. The memory and storage is completely on BYOC setup and can work with confidential workloads
I've always been interested in health, longevity, and biohacking.
I'm Canadian but I moved to USA to work for Microsoft. One thing I liked about the USA was how easy it was to order bloodwork/lab testing. My original idea was to bring that experience to Canada, but since starting I've found other ways to improve the patient healthcare experience in Canada too.
My company is https://teletest.ca
I'm reverse engineering the streaming protocol for three different brands of cheap surveillance video cameras. Each brand has its own app. They're all awful. None of them have a native desktop MacOS app. I've successfully created a native app for MacOS and Android for one of the brands. They're faster and work better than the ones they're replacing. One down, two to go.
I'm working on Hale, an abstract distributed systems language.
It's a traditional systems language in the sense that you have fine control over how resources are used, but it's abstract in the sense that you have to come to an agreement with the compiler about your intent around the resources, rather than directly declaring the machinery.
It's a distributed language in the sense that there is a higher level structural abstraction in the language that situates computations of the program within a recursive graph structure. Part of that structure incorporates message broker semantics. This gives the compiler a unified view of programs across arbitrary process boundaries. The structural abstractions also enable an effects+formal verification syntax within the language, which allows programmers to codify architectural decisions that are then checked by the compiler.
The compiler itself is infrastructure aware, and the highest layer of functionality it handles is configuring the environment into which applications deploy. The single language handles everything from individual memory allocations to distributed fleet deployments.
The language is close to v1, and I'm currently building a fully distributed inference-as-a-service application with it. Arbitrary hardware/supply matched against arbitrary callers/demand in real time.
https://github.com/hale-lang/hale https://github.com/hale-lang/voice
For the past two years I've been working on my day planner app Ember.
Most planners treat your schedule as fixed, so the moment a meeting runs long or a task takes twice as long, the plan is wrong and you either fix it by hand or give up on it. In Ember, if you move a task or change how long it takes, everything after it shifts later, and it flows around fixed things like meetings. It's been super helpful and I have been adding new features ever since.
Right now I'm working on a native Kotlin app to replace the current one is web + Capacitor and exploring full duplex voice models so I can add or move things around the app while having a conversation with an agent.
homepage: https://emberflow.space demo: https://demo.emberflow.space/?ref=hn
Still plugging away on Flopper to build a holistic view of the GPU market including providers, rental prices, vendor information and mostly - specs for GPUs.
Added a way to look at power/flops and new model requirements lately to try and estimate how many GPUs you need to run a certain model: https://flopper.io/tools/llm-gpu-calculator?model=llama-3-3-...
When I was a kid I loved a game called "Stars!" from 1995. I cloned the core and I'm also adding my own touch and lots of high sci-fi concepts. Already fun to play, early access coming soon: https://store.steampowered.com/app/5482440/Asteria_4X
For now Im working on an agent-orchestration application that is taking tickets from a tracker and giving them to an Agent. It's called Even Keel.
The main feature is, that I can use different subscriptions/API keys (Claude and Codex - v0.1.4) and trackers (GitHub, Linear and Huly - v0.1.4) to make Even Keel start an agent for every single issue, task by task, one at a time per folder. There is also a parallel feature, so I can run more than 1 project at a time.
Once upon a time, when codex agent was integrated into the app, I wrote a ticket to make codex his own hook. It reused a hook that blocked its own git merge, if tests did not pass. I started calling it "A new agent writes its own cage"... Right after, it turned out it doesn't listen to the hook and it fixed it
Version 0.1.4 is already shipped, but v0.2 is coming soon with more supported trackers, agents and new features.
A video game in which players can build their own worlds. The brains of the creatures in the video game are a small language model + harness that runs in your browser: http://id-le.com
It’s open source: https://github.com/nicobrenner/id-le
I’ve been working on my inertial navigation library. It actually beats ArduPilot’s navigation subsystem in my benchmarks. I’ve been working on these algorithms since my PhD thesis and now I am releasing the stuff as open source embedded C and Python library:
Still changing along on Incredible (https://www.incredible.rs/) a Rust Text User Interface (TUI) Framework for the 2nd Quarter of the 21st Century.
It is New, Experimental, Modular, Declarative, Event Driven and Reactive. It builds cross-platform apps that can be compiled for the terminal (macOs, Windows, Linux), a standalone native window (macOS & Windows 11), or the browser (via WASM) from one unified code base.
Curious if this project will ever be 90% done.. ;)
I'm working on an AI powered loan servicing platform. We've been trying to get cold / warm intros to onboard a set of design partners
How to make data storage on multiple storage devices unbreakable and consistent? The former means for me that back-ups should be automatical and serverless, the latter means that no matter how much of that HDD/SDD devices are connected, I want always to know all metadata of all files I have. I want this software under GPLv3 if to consider GPL still protects me from anything because permissive licenses are no better than WTFPL by any single case.
My first idea is to require the user to have more than one disk for the system needs and the system should know all of the rest data devices user owns. My second idea is to treat different files as having different priority which impose multiple copies. My third idea is to make the software accessible, and the last idea is to make encrypting of all the devices easy and with no ties to any commertial entities. I would like to have a discussion about how to not lose anything digital.
HostsTray, tray utility to manage your hosts file, that also has a reverse HTTPS proxy builtin, that installs a self-signed CA and generates certs on-the-fly.
https://github.com/maxmart/hoststray
Also it forces your browser to "flush" their DNS by killing the keep-alive connections that for example Chrome keeps open in background. And it can be controlled by your local coding agent through a simple API (which also serves the skill for it)
Still cracking on with https://whistle-enterprise.com and making improvements to the document generation model. I'm ditching my old trusty Phi4 model and testing out some other small models to fine tune, something about Phi4 just seemed to work quite well for what it was so it's a hard contender to beat for midrange laptop CPU inference at a decent rate and quality.
I’ve been working on Beacon, which just got Beacon approved on the App Store! It’s a mobile app to instantly answer “which friend is free to talk right now?” and summon them in a single tap.
Send your beacon to a group, and everyone gets a gentle ping. The first person who answers gets connected for a 1-1 call, and for everyone else the signal drops silently. It does not log any missed calls so there’s neither pressure to answer nor pressure to return calls.
Overall it works really well! It’s best if you're able to join with 4+ friends/family you’d like to chat with more. I have dozens of connections in the app now and it feels like absolute magic to me. I tap the button to send my beacon to everyone and get a random friend on the line.
I wrote a children’s story in Galician about open source. It's told as a fable. I made it open source, and people are now helping translate it into Spanish and English.
I’m looking for ways to fund printing and distribute the book. If anyone has done something similar, I’d love to learn from your experience!
Just released this: https://github.com/rlorenzo/deadwood-rs
Since everyone is rewriting everything into Rust I created a tool akin to fallow (https://fallow.tools), but for Rust:
Finds dead code that LLMs leave behind.
>Making a simple game
>Want player scripts to be possible (And dogfood it for enemy ai too)
>Behavior tree dep looks too easy, let's make a dsl + interpreter
>Typed builders are annoying (because stateful), lets separate into builders -> untyped ast -> typed ir
>Hm, typed ir is super repetitive, now I have a schema (ts) -> codegen (ts) -> .ts / .rs build step
Fun waste of 2 weeks, not even counting the time spent on pl literature
Working on a tool to provide a thin OIDC auth layer for self-hosted LLM engines. Users log in with your Org's identity provider, generate an API key for themselves, point their harnesses at your instance of the relay, and then they're off to the races.
No web chat interface, no call-us-for-pricing feature gating, just a tight Erlang program for people who want to get shit done!
I vibe-made a Russian life simulator called Frontline Rush:
I vibe-proofed the ABC conjecture (IUT LEAN Fable):
https://github.com/oneman/motizukipdf and https://1234567890.help/
Yeah
A networking-over-DNS client and server, in Go. https://sr.ht/~nullenenenen/ShrewDNS/
A bit like iodine in the basic idea, but the execution and network topologies you build with it are different. It's more of a building block; you get your DNS channels represented as UDP endpoints, and take it from there, ie you can run Wireguard over it, or SSH or a SOCKS5 proxy (using quicmux, so over QUIC), or run Mosh straight over DNS if you're crafty.
It has an Android app too! I tried to submit it to F-Droid but they're so swamped in spammy or low-quality apps that they're not taking any brand new apps :-( A sad state of affairs, that.
For the first time in along while, I'm not actually working on anything. I do some vibe coding for fun.
I'm kinda leaning towards vibe coding my own GUI coding agent. I currently use Pi and recently have been playing with Herdr and Paseo, but none of these feels right to me. These are amazing tools, but ai think we could do better. But I'm very hesitant - I know the only way to "succeed" in a project like this is to write (vibe code) hundreds of thousands of lines of code and it's kinda depressing.
An easy way to find a place that works for everyone. Think of it as a doodle for places.
You can also create collections of places, ask for recommendations etc.
Working on a Chrome (& Firefox) extension to make finding and downloading Free DLs on Soundcloud easier.
Artists make users jump through hoops to like, comment, repost etc.. before giving them access to the download - CrateDiggr highlights the free downloads in the Soundcloud UI and clicks through the download gates for users.
Been a fun passion project, check it out if you download tracks from Soundcloud!
Getting https://streamish.github.io/ ready to launch, a music server compatible with multiple 3rd party mobile apps and particularly Synology and QNAP. Want to do Plex next but I suspect that's a dead-end since I think you can't directly sign in you have to do it through their website.
Now putting it to personal use and keeping an eye out for issues and rough bits to smooth over.
Built with -
NestJS, Sequelize + SQLite, OpenAPI, and Jest
React, TanStack, Shadcn, and Playwright
Built an app to suggest best local beach for given weather for my town. My partner shared with local moms and they seem to be loving it - https://beach.metacursion.com/
Also revived old project - https://aminosplore.metacursion.com - the idea was to compare what essential amino acids each food contains using spider web chart. It’s been dead for a while due to USDA API changes so I revived, added macros and enabled to build foods into actual recipes.
Finally have been improving elsrc.com - an easy and scalable way to use ai image generators. If you need hundreds of images you can just request them by url. Supports versioning, batching, regional and national styling based on viewers IP and few other neat features!
I've been working on a planner for the past 2 years for people whose day never goes the way they planned it.
I was trying to solve the problem I had with time-blocking, where your day ends up derailed and you give up on it because it takes too long to replan. I have solved with the idea of a cascading calendar. If I move a task down or change its duration it pushes other tasks later in the day.
Since then its kinda grown into my full productivity system for logging my thoughts and planning and documenting my life
You can demo here for free https://demo.emberflow.space/?ref=hn
https://www.haptrix.com - An Core Haptics editor, with the release of the iPhone Duo, the editor finally makes sense on device, so I've been working on the new UI.
https://www.phonique.com - An AVAudioEngine/PHASE Audio graph editor, it's easy to connect nodes together visually, then export the wired graph as Swift
A small team and I are working on Sendery [1], a transactional email platform for product teams. We're trying to fix some of the headaches that come with managing product emails as companies scale: keeping track of hundreds of templates, maintaining consistent designs, coordinating changes between developers and marketing/ops, and managing everything across multiple languages is way more time consuming than it should be.
It's been a fun project to work on since it's been a pain point for myself on many projects, so I'm mostly just trying to solve my own set of problems and hoping other people find it useful too.
https://hn.fiodorov.es AI-powered search across all HackerNews comments
https://climate.fiodorov.es Agentic analysis of climates of world cities with more than 300k inhabitants
Both have mcp.
https://github.com/a-poor/ganymede
I'm working on a native Jupyter client, written in Rust!
I've been working on https://gitsocial.org/ for a year and a half now. The approach is to keep it all in git: issues, PRs, comments, releases, etc.
Hoping it becomes a viable self-hosted alternative to GitHub, GitLab, and Codeberg!
https://github.com/ivankovic/stop-bots
Trying to keep my selfhosted server safe with a Rust TUI for managing bot blocking at the nftables and nginx level.
AGPL v3
Still working on ShopSpec, a parametric deterministic furniture builder. Trying to find time to put more actual builds in the gallery - have a few projects that are in progress while continuing to build out the tool and iterate where I find personally useful.
Hence (https://hencegame.com) is a modern twist on Einstein's puzzle, with fun picture clues instead of text. It's a free online daily puzzle, the same for everyone, where every puzzle is always solvable by pure deduction with no guessing. I launched it October 5. The hardest part was the graphics, making each picture clue unique enough that you recognize it immediately and know what category it's from, even at small sizes. Easier said than done.
Every puzzle is run through a verifier first to make sure it is solvable without any guessing, just by pure logic, so there is always a next step to find, which is what makes it feel so rewarding. There's a hint system to help and it works really well if I do say so myself . No account needed to play.
From what I've been hearing so far, the people who have checked it out are enjoying it and can't wait for tomorrow's puzzle, so that's a good sign. I'd love to hear how the difficulty feels, and whether the pictures read clearly.
I'm still working on my own shell based AI tool. This month I am testing and reworking it to try and get useful results with really small models, like 2B level. Goal is to get help debugging Linux issues when offline with these small models that can run anywhere. Then also being able to access cloud based models to do harder stuff once online. all while feeling like a CLI tool.
github.com/patrickjh/ssa
I just got Beacon approved for the Apple App Store, it’s a mobile app for answering “which friend is free to talk right now?”
It’s a one-to-many phone call that only one person can answer. Send a beacon to a group, and everyone gets pinged at the same time. The first person who answers gets connected for a 1-1 call, and for everyone else the signal drops silently. It does not log any missed calls so there’s neither pressure to answer nor pressure to return calls.
Overall it works really well! It works best if you're able to join with at 4+ friends/family you’d like to chat with more. I have dozens of connections in the app now and it feels like magic to me. I press the button and get a random friend!
An MIT re-implementation of an AGPL project with a very anti AI community.
I’ve been working on an accounting agent factory (https://www.translucent.io) . Building a harness and infra that can process books with thousands of invoices, transactions, etc. is quite rewarding and challenging. In particular the long-running nature of things.
I'm creating an IDE for founders to run their operations. Create automations (deterministic), dashboards, and soon notification systems. It's a desktop app, and the whole IDE runs in pair with claude/codex through local mcp.
It was developed as an internal tool at my other startup, Ipek helped us to run the whole thing smoother and with all the context I need.
https://ipek.run