alt Hacker NewsAsk HN: What Are You Working On? (July 2026)
What are you working on? Any new ideas that you're thinking about?
Comments
PastML - https://pastml.com/
AI-first, MCP ready to host single HTML page. Connect & publish directly from ChatGPT app.
I finally decided to put together a Sonos controller with the navigation I wanted and SMAPI servers for the live music archive, and all the grateful dead and phish shows. Thanks Claude! A PWA with tailscale and I have a controller that does what I want and works at home on an S1 system and at the beach on an S2 - seamlessly. Better than the "real" thing as far as I can tell.
Working on a no-code tool to make it easier to self host fully compliant databases with an audit first approach.
Startups should be able to deploy their own SOC2 compliance ready databases without paying cloud providers to do the deployment for them.
Why pay $1050/month on AWS RDS when a Herzner dedicated server costs $55 a month and offers more resources?
AnswerJournal lets you save AI answers to a personal journal just by saying "save that to my AnswerJournal" mid-conversation.
Each answer gets its own URL, and posts can be public or private. It's a bit like GitHub meets Stack Overflow for the answers AI gives you.
It connects to any MCP-compatible AI (ChatGPT, Claude, Grok, etc.) through one server URL.
https://flights.benjaminbenben.com/
I’ve been playing around with aligning drone footage to flight paths. I'm really interested in the idea of representing a video as a volume, planning to do something similar with non-drone video too.
Just a personal project right now but it's a music metadata aggregator.
I find that nobody really knows how to do this. Machine learning can detect some song attributes well (bpm, ez right?) but it's inconsistent with some things (eg mood, spotify valence)
I prefer to only add metadata that I can rely on: track credits & instruments (when available), lyrics, bpm / "energy" and genre. At least that's what I've got for now. I'm not adding anything unreliable.
So far I'm able to pick a genre, artist or even better, song and it gives me a list of tracks that are similar. I can alter the weights of "era", "instruments", "genre".
So far i haven't run old school NLP on the lyrics but that's the next step. It's likely to be far more informative than "valence"
Anyway, not public, still very alpha but I like it and find it useful.
Recently launched https://hellbox.com (formerly Font Proof) and am still actively working on it. It is a native macOS app for font proofing. It watches your type design software like Glyphs, RoboFont, and FontLab or any font file or designspace on your machine and reloads the font proof when you save.
I'm working on dotUI. A framework for developing UI's where the core logic such as routing, events, state management is defined by the user and the llm handles the styling. This way every user can have their own UI's without losing functionality. Will post here regarding it in the coming weeks
3D model construction of climbing gym. There is so much detail a normal climbing video can't show. Wall angle, hip position. I'm hoping to overlay human model, taken from climbing video, onto the 3D wall to make climbing visualization easier.
I’m working on a ranking system that combines my running distance with my total commits. Mostly to motivate myself to be productive and healthy, and also just to have fun. I integrated it into my GitHub profile: https://github.com/Noc2 Code: https://github.com/Noc2/paceandpush Try it (only iOS beta): https://paceandpush.com/
I recently started working on SnapSense, a macOS app that OCRs your screenshots and uses a local LLM to auto-rename them into something readable.
https://github.com/Enigma-52/SnapSense
This was solely built after opening the wrong 'Screenshot 2026-06-03 at 6.09.28 PM.png' for the 4th time :|
I made a tool that creates sandboxes (docker, podman, orbstack, seatbelt, tart, Apple containers, containerd, kata, firecracker) and then sets up an agent (claude, codex, gemini, aider, opencode) inside it with max permissiveness (no prompts to call sed, etc).
It creates a CoW copy of your workdir for the agent to play in, and then you pull changes out using git diff/apply semantics.
You control network access, secrets, which files/dirs it has access to.
It's a MASSIVE time saver, and I use it as my daily driver.
I am currently focused on an educational site https://www.higherscoresdfs.com/dfs/spa/welcome/ that helps Daily Fantasy Sports fans improve their DraftKings and FanDuel gameplay. We offer strategy advice, player tips, optimized lineups, and player props. We now support both classic and showdown games and have recently added a section offering game outcome predictions for those who do sports trading on prediction markets. We currently support the NFL, NBA, and MLB leagues.
I made an HN filter for substacks and blogs: https://hnsubstacks.com/
It allowed me to explore a serverless deployement (on CF workers) with a toy project, that I wanted to make for myself.
Repo here: https://github.com/ariroffe/hnsubstacks/
Today I'm carrying on looking for a literary agent to represent me for my young adult speculative fiction. Long after human civ collapsed, crows have evolved to speak and are undergoing their own industrial revolution when they discover an asteroid heading towards earth. My 10 year old read it for the first time yesterday and loved it. That's made it all worth it already even if it doesn't get published.
Tonight is a meeting of a local group growing hemp locally, processing and spinning by hand to make an item of clothing locally and sustainably between us all.
Just spent the weekend at a wing chin gathering with some incredible people. They showed me so much I need at least 2 weeks to think about it all.
Apart from that I’m starting college to study therapy and counselling soon so I'm trying to read up to be a bit prepared.
I'm working on improving my Indiekit https://getindiekit.com (not mine, just a big fan & heavy user, to which I created plugins : https://rmendes.net/changelog/ and also working on a local Ai-stack deployment so that I can put that DGX Spark for a good use !
A self-hostable Claude Tag or OAI Work [0].
A while ago, I realized that most new agent harnesses being built must be hosted on your machine or on a VM--in other words the agent needs a full OS process at all times.
But we do not have good harnesses being built that are multi-tenant, do not use compute while they are paused, but are are still as powerful as, say, Claude Code or Codex, OpenClaw.
So I set out to build one. I realized that the best substrate for these kind of agents are durable workflow engines. I'm currently supporting Temporal. AFAIK much of OAI agent infra is built on Temporal too. My harness is decidedly not just another agent SDK, but rather a battery-included product.
I'm working on a neovim plugin: https://github.com/patrick-sharp/waypoint.nvim
It's a bookmark plugin with a couple features I felt were missing from existing bookmark plugins. At my job I often have to navigate large codebases that I didn't write, and it's easy to get lost. This plugin should help with that.
Being unemployed and wanting to make something, I started studying quantitative trading concepts and got into algorithmic trading.
I decided to build out an algorithmic trading platform using the tools I developed for myself.
It's written entirely in Rust but user algorithms are written in TypeScript. It uses a Cloudflare-workers inspired approach to run the user functions.
The server uses under a megabyte of ram to run and user functions also use a negligible amount of memory per invocation.
It's also super fast, with round trip latency of 3ms - well, at least it does when I use the proper server. I'm running it on my low cost server right now so latency is around 50ms.
I know no one will use it, but it's been very fun to make
https://trailer.dev, a self-hostable development environment platform.
Been working on it on and off for 4 years now. I used to lead an ML consultancy and one thing I've learned is that data scientist DO NOT understand reverse proxies, nvidia drivers, docker and can not set up complex environments for their projects, especially ones that would support collaborative work.
It's been really fun to work on this, currently working on Kubernetes support.
Two weeks ago presented current state of KEIBIDROP at Pass The SALT 2026, and now I am planning the push for the next version 0.4.0.
KEIBIDROP: Makes remote files appear as local (it hides the network latency in order to let you open and edit a peers file without downloading it upfront or re-uploading it fully back).
For version 0.4.0 I am planning multi-user support, using UDP (QUIC) instead of TCP as the networking layer, optimization of live-edit regions of files, and to test it even more for data heavy workflows.
Here is the website: https://keibidrop.com/
And here is the github: https://github.com/KeibiSoft/KeibiDrop/
I've been building a text editor. It uses a server-client model, with native/terminal/web clients using a shared (local) server instance. It integrates LSP/tree-sitter/ripgrep/Git/etc, and has a concept of 'workspaces' to easily switch between projects.
I spend less time in an editor now, but it's been satisfying being able to take features I've enjoyed from other editors and customise - and it's oriented a bit more towards exploring code than editing. Key-bindings are arranged such that bare letters are for motions, Shift- variants are then used to extend selections, Ctrl- bindings are for buffer mutations, and leader-prefixed bindings are application level (e.g., opening pickers).
I’m still working on a price API for the Counter-Strike 2 market that provides real-time & historical data.
Since my last post in February, I’ve gotten to ~25 paying users, which is cool considering it started as a fun project. Sorta a niche within a niche here.
The market is distributed across a bunch of 3rd-party marketplaces, and there's no 'simple' API that provides genuinely high-quality data for the few marketplaces that matter. It’s a surprisingly complex problem, which is probably why nobody else is bothering :).
It's been a super fun project, and I've been able to learn about collecting & managing a high (to me) scale of data, building an API from the ground-up, and creating my first 'commercial' website.
Website is @ https://cs2.sh/
The API is built w/ Go & Clickhouse, which I've also been super impressed with so far in terms of performance and efficiency.
Web design is inspired/somewhat taken from turbopuffer's site, since I really liked it.
Still doing https://rcarmo.github.io/projects/piclaw (which is my main “IDE” now), ended up forking https://github.com/rcarmo/tau-prime to have something I could run on my iPad (and on the Mac, with a simple sandbox), hoping to get back to 3D printing (https://taoofmac.com/space/notes/2026/07/12/1230) and playing with hardware.
Oh, and migrating most of my stuff to microvms on https://rcarmo.github.io/projects/pve-microvm/
I've been working on MacDispatch https://macdispatch.jurajb.dev/
A macOS menu bar app to alert when apps start using too much resources and drain your battery. Helped me diagnose many leaking Chrome tabs and macOS bugged services.
https://aportee.ca - interactive walkability maps for Canada where you can pick and mix the amenities that are important to you.
I'm working on a technical interview service that brings up a real production environment and also lets users use agents and our system tells the interviewer how the candidate really did.
Also, in this environment, users can have a full cluster with Kubernetes, Docker, or normal Linux, with a database and all sorts of dependencies.
Please check it out and let me know what you think: https://easyenv.io/
I'm working on a side project to clean up and fix EPUB files with transcription errors, incomplete or inaccurate metadata, and other issues. Users can accept/reject/edit the fixes as they go through it. I have the command-line interface and TUI working (thanks to Ratatui), but I'm still working on a Calibre plugin and web-based version using WASM.
Building a new Smalltalk VM from scratch that better utilizes modern hardware (full multicore support) and a web-based system browser so I can develop with it remotely.
I'm working on https://gatewai.studio in my free time - it is an agentic workflow engine for content creators.
There are 40+ nodes that can be used to generate and modify images, videos, audio, or vector graphics. Some of them include Crop, Resize, LUT extraction, Levels, Audio Compressor, Ken Burns, Mesh Warp, Recorder, Noise Gate, Compositor and Signal Builder.
It also supports signals for dynamic and time-based configuration values for the nodes. For example, making blur strength change from 30 to 0 gradually in the first 2 seconds of a video.
It uses a WebGPU pipeline for rendering and a homebaked engine for workflow processing.
It is free to use except for the AI nodes and workflow agent. It is not officially released yet, and feedback would be very valuable.
Building a rootless, namespace powered (deeply stretching the definition of a container), on demand application workspace.
- Each component in a mini app in a heavily locked down container - Components are deployed and built in a web workspace, in the same workspace you can open a terminal and use your favourite coding agent to work on component code (each terminal is itself heavily sandboxes, has rw access only to the edited component code and users home dir) - Everything comes with heavy rbac and minimum permissions - Oh so much more
Explaining this well is hard, much like explaining to someone what Kubernetes or AWS does. This is at a level of what a sophisticated company infrastructure team would run, just as a workspace you can deploy for yourself easily and agents just build within that framework (I’m a cofounder of a infra/compute/datacenter startup and intimately familiar with this kind of complexity)
The main thesis is that Claw-style agents still feel like school projects, and that in the agentic era apps on demand will be more of a thing, and that the current systems weren’t built to deal with a whole new app built every few minutes.
May or may not end up as open source soon
https://finbodhi.com — It's an app for your financial journey. It helps you track, understand, benchmark and plan your finances - with double-entry accounting. You own your financial data. It’s local-first, syncs across devices, and everything’s encrypted in transit (we do have your email for subscription tracking and analytics). Supports multiple-accounts (track as a family or even as an advisor), multi-currency, a custom sheet/calculator to operate on your accounts (calculate taxes etc) and much more. Supports price for most Indian investment vehicles and US stocks.
Most recently we added support for creating custom dashboards. You can compare return with leading/trailing/rolling charts for investment options and benchmark (create custom dashboards tracking nav and value chart of) your portfolio (or a subset of assets you own) and US stocks, etfs etc. And family dashboard (e.g. you can see networth, cashflows, income, use sheets at family level and more). See https://finbodhi.com/changelog for details.
We also write about related topics:
We wrote about comparing investment options: https://finbodhi.com/docs/blog/compare-charts
Benchmarking your returns: https://finbodhi.com/docs/blog/benchmark-scenarios
Understanding double entry account: https://finbodhi.com/docs/understanding-double-entry
I started burning down the backlog of all the stuff I wanted to get to for side projects but never had time for (before LLMs):
- https://smacke.net/ffsubsync -- automagically synchronize subtitles, now purely client-side in your browser thanks to pyodide
- https://ipyflow.github.io/ipyflow/lab/index.html?path=demo.i... -- reactive python jupyter notebooks, again in the browser thanks to pyodide / jupyterlite
- https://smacke.net/pipescript/lab/index.html?path=demo.ipynb -- magritter-like pipe / placeholder syntax for ipython / jupyter, again able to run purely in the browser
- https://smacke.net/pycograd/lab/index.html?path=pycograd_sim... -- pyccolo and pipescript-powered autograd, once again able to run purely in the browser since numpy has a wasm target (notice a theme here :) )
Policy guardrails for Coding Agents. Never babysit your agents - github.com/lud1161/agentjail
· OS native sandbox (<4ms start-time) · Policies for AWS, local files, k8s, DBs etc. · Rego backed policies w/ OPA (CNCF Graduated project) · Fastest policy evaluator (<3ms)
Please star the repo if you like the work.
Building an unified admin platform[0] for Source game servers & the alikes. I grew up playing Source games and ran a few, but there were always gaps between all the various OSS plugins you install and were quite fragmented, and never designed to work well with each other.
I am working on a Jupyter notebook client for VisionOS. It allows for 3D data to be visualized on visionOS. Right now it supports point cloud data, USDZ models and Gaussian Splats. I am working on it to launch on the App Store. Sign up for more information at http://www.pulto.org
I rewrote Chucks [American] Football Pool in Newspeak!
Having a hard deadline of Sept 1st, I wasn't sure three months would be enough time. In Newspeak, it took all of three WEEKs to get it to it's current state.
One of the best features is the fact that it's local ONLY. It uses IndexedDB (Newspeak library written by yours truly) to persist the data, meaning zero backend headaches. It also makes it usable by the masses where the Seaside version had many problems in this regard.
Chrome only. No Clod.
https://chalculator.com/primordialsoup.html?snapshot=ChucksF...
AI resume spam has absolutely ruined hiring. I've seen countless stories of hiring managers complaining about receiving thousands of resumes within an hour of opening up a new job listing. Likewise, good qualified candidates are getting drowned out by loads of customized junk. I'm building a tool for the recruiters and hiring teams to combat this and bring some sanity back to the already challenging process. The general idea is to introduce a small bit of friction that will hopefully be not-too-impacting for legit applicants while stopping bots from proceeding.
First iteration is ready to fly, just working out the infrastructure at the moment. Hoping to drop this on Show HN soon. If anyone is interested in test driving this prior to launch, I've temporarily added my email to my profile.
I’ve been exploring what sort of agentic tooling to write for creative coding and realtime VFX. My second iteration just got released earlier this week (also open source): https://sxp.studio/apps/subz
If you’re open to the idea of composing code blocks and ideas, plus some generative UI exploration, feel free to join!
In the past few weeks, experimenting with a custom ai agent running entirely in the browser with sandboxed tools exec.
The core is built in Rust, a native CLI is built on top for local experimentation but the most interesting part is the web version: the core is built to WASM and get augmented with many tools in the JS land: - OPFS access (read, list, edit files) - Sandbox Python exec (Pyodide in WASM) - Sandbox DuckDB exec (DuckDB-WASM) - Draw charts - Show images - etc OpenAI Completions compatible API providers are supported.
But if you want a full local and sandboxed execution of the whole agent, the web version bundles also wllama to serve local GGUF models (with WebGPU optional support).
Github repo: https://github.com/rclement/cooper
I’m trying to build the best TTRPG chronicler: https://loracle.app
Everyone who plays D&D has experienced the moment where they forget key details about the collective story they’re building. From ‘hey it’s been a month, where are we?’ to ‘wait who was this crazy npc again?’, ai is excellent at transcribing, notetaking and building a knowledge graph of your fantasy world.
I’m still building mostly for myself by adding a ton of features I know my friends would want, but also think there’s some ‘there’ there.
The idea is simple: let Loracle record your sessions on discord or upload the raw audio of your sessions, then get a rich personal wiki and session notes you can interact with.
If you’re mid-campaign you can also upload session notes from plain text and it bootstraps a campaign wiki. Then future audio based sessions have a good base of npcs, quests, characters, etc to build off of.
At this stage I’d love feedback more than anything else. Happy to comp a lot of usage to HNers in return for some reports on how well it’s serving you. Email [email protected] for anything and everything.
I've been automating many chores that i find myself usually doing with email: managing calendar invites, publishing and sharing calendars, but also want it to act like a proxy to avoid giving away my actual email in different services. It's a work in progress but I love cloudflare ai gateway, been using it to bring some ai into the functionalities. Future things: handling newsletters, more ai free use (?)
If you wanna test it, please ping me so I add your email to allowlist!
A text-based song format for generating music. I wanted to be able to create a song entirely using text, so I created a TOML-based format for doing so, and gave it most of the features you would find in a DAW. Since the format can be described in a SKILL file, AI can be used to write a song in this format, which can then be converted to audio.
I'm building https://mallcop.app, an AI, open-source, cybersecurity monitoring, investigation, and escalation solution. It's rooted in four principles:
1. Analyze data where it resides
2. Connectors and detectors are not a moat
3. MIT license every line of code
4. Fully exploit AI for investigation, maintenance, and self-improvement
MallCop runs on Donuts (e.g. tokens), which I'll happily resell, or you can BYOK and use your inference API of choice.
I'm working on a no upload, no account, no server-side preprocessing online viewer for schematics, PCB layouts, 3D boards and BOMs called ECAD Forge -> https://ecadforge.app/
Building out a housing heatmap of England and Wales (Scotland and Northern Ireland publish housing data differently, working on getting them integrated!
I was interested in learning more about reverse image search, so I built audioguide.london. It’s a simple image search where you take a photo of an artwork in a gallery, tries to ID it and returns a pre-generated audio and a link to the museum website.
The audios were all generated locally, essentially looking at the contents of the website, running it through a LLM to generate a script and Kokoro for TTS.
I’ve built it as an app for myself almost a year ago, so I deployed it as a vibe coded website in here: https://audioguide.london/
I'm working on the finishing touches for a big new "Event Filters" feature for my Shopify app, Stages (https://getstages.com). The feature will let users set up rules to decide which orders should be imported into the app based on certain criteria like Shopify product names, collection names, order value, and so on. Users have been asking for it forever, and I'm planning on publishing it this week!
I'm also working on an update to ShopifySharp, the .NET package I maintain for Shopify's graphql and rest APIs. I need to regenerate the graphql types and the fluent query builders for the July 2026 API version that was just released, and I'm planning on some extra QoL improvements that I've run into while using the package over the last couple of months. I particularly want to add some F# QoL features, since I wrote the package in C# but use F# in all my personal projects. (https://github.com/nozzlegear/shopifysharp)
I’m working on a NotebookLM competitor, sort of an open-notebook in the cloud with a lot of features I’ve added (and made available as either PRs back to the base repo or public fork) such as a plugin system to add more generated “creations” or a quick way to spin up your own compatible cloudflare workers ai setup to test out models. I’m trying to incorporate the same bookmarking features I love so much with Raindrop.io (big fan) and experimenting with a new wiki generation feature. It’s only soft launched but happy for any feedback. https://notebooker.ai
I'm building a free browser based hub with a variety of apps that I find useful for me.
For now, I have 8 apps and I'd be grateful if any of you could give me your impressions.
I'd be happy too if any of them could be helpful for you.
https://dailypocketapps.com