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
I'm working on the Obsidian plugin for building and rendering family graphs [1]. I started my own genealogy research in 2025 and did not want to store data on a 3rd-party platform. Obsidian was a great decision. The only thing I needed is a tree (graph) viewer for family members relationships (something like myheritage.com has). So, I built my own with a lot of interactivity on top.
https://trackdayz.io - motorbike track day calendar and social hub
Currently in the early access and bootstrapping phase, the system is meant to help you find which event you can go next and also plan your whole season, organize your calendar, link up with your friends, track your progress as a rider, and see where you stand on the global rankings and between your buddies.
There are more ideas than time to implement them as this is purely a hobby project, but doing my best as I go along. Planning to start advertising it a little bit in relevant groups in the coming weeks.
I've been building some sqlite plugins for playing with ngrams for text search. I'm not sure why, but I've learned a lot about the internal sqlite apis and it brings me a lot of joy. I would like to start a blog detailing some of this work but haven't found the time yet
I've been working on Next Move Theory for the last 8 years.
Next Move Theory is a methodology with a step-by-step algorithm for every product decision. It lays out every tactical and strategic move open to you and helps you choose the best, with the odds on your side. The foundations are open and free. AI skills run it on your product.
Dozens of cases in my home country, zero presence in the US — now I'm working to make founders and PMs in the US aware of it.
1. A compiler for real-time tensor processing (arbitrary DSP, ML). In something like LISP or Haskell, the goal is to compile lambda calculus for fast/reliable execution—as such, you can express a program in a fully general language that can represent any computation and execute it without explicitly modeling the lower levels of the machine. I'm building a compiler that does the same thing for the subset of programs that are guaranteed to execute on-budget. The effect: you write code that looks like DSP/ML math and it compiles/runs optimally with execution guaranteed by construction.
2. My take on an agent framework ... append only log + content hypergraph in Elixir, tools that regularly pull data from other services into Postgres—built as a kind of 'exoskeleton' around claude/codex so it's not competing with fast-moving tools.
Thinking about category theoretic models of computation: https://arxiv.org/abs/2208.03817
--
Some things I want other people to build:
- Indexing for Github
- All-in-one social media ingestion libraries for agents
- GOFAI-inspired knowledge / semantic / research graph stuff—I want to point agents at rules/structures for writing connected, verifiable statements
Making a little static site to document the AV setup at my church.
Started with just an html page with buttons for shortcuts. Added a few paragraphs and then got tired of writing html. Not many simple wysiwyg desktop html editors surprisingly.
Then discovered Zola, a static site generator that takes md files as content, and adds them to template html files.
Perfect for my little use case. I'm having fun with it :D
A backyard wildlife narration that takes video clips from different styles of cameras and turns it into a documentary style narration - https://everydayearth.ai.
It originated from a project with my son where we created a nesting box out of an old wooden shelf and added a camera on the inside to see what happens. It was taken over by a Screech Owl and it's been fun to see what happens in our backyard at night.
I'm learning Rust, and a bit about audio processing: https://github.com/leocrabe225/audiodsp Just before that I was learning TypeScript, and I'll be coming back to that project as it's quite the process. I'm leaning heavy TDD and rely on the type systems a lot. The project is called "Rentier", a joke about the Monopoly board game. https://github.com/leocrabe225/rentier-rewrite I finished part 1, with a postmortem. Part 2 is React frontend. Part 3 is implement the postmortem findings in the TS engine. Part 4 is update the React frontend for the updated engine. Part 5 is make the engine in Rust from scratch. Part 6 is point the React frontend to Rust engine, and praise the pros of having a good architecture. Kind of an ode to clean code if you will (Many things to learn along the way).
I've been working on a home Hermes setup that I'm using to do a variety of things. So far, here's what I've done:
- Set up an in-house alternative to my Ring subscription for the cameras around my house. So far I have it real-time monitoring, complete with AI face recognition and interpretation of events, all on an internal web page. It also sends me alerts via Telegram.
- Set up a blog that I'm using to catalog these experiences with Hermes and AI in general
- Started working with multiple agents to do things when appropriate. My main model right now is glm-5.2 for cloud, and Qwen3.6b-IQ4 (4-bit quantized) running locally. It only takes 18GB of VRAM on my 4090, so I have plenty of overhead. I'm also using Hindsight instead of the MEMORY.md that Hermes natively uses.
- Setup local image generation, with the local Qwen model, using ComfyUI w/Flux.
- All of the above isn't including the numerous smaller jobs (like setting up Telegram, setting up automated cloud backups, troubleshooting Linux issues, etc) that I've been using Hermes for.
Future plans: I'm working on making a game with Godot, learning as I go. I haven't had Hermes do anything for that, and I don't want to really, except I may use local image gen for testing purposes, as I plan to engage an artist for any graphical work in the final product.
I'm doing all of this just a learning experience. It's been really fun so far.
I'm exploring triply periodic but non-minimal surfaces. https://codeberg.org/okaleniuk/faitps
Triply periodic minimal surfaces are the golden standard in thermal management, acoustics, and even medical applications. But minimality itself doesn't contribute much to practicality. We use them because they are simply studied better than the non-minimal surfaces.
So I'm studying the non-minimals. They are much more governable, what I link to is a demo of a surface builder with two levels of control. Next, they are conjugatable including conjugations with different period of self (that will be the following paper), they generalize nicely to non-periodic or partially periodic surfaces, and they work in other space configurations. E. g. I'm now playing with bi-periodic curves that cover the 2D space with self-replicating hexagons.
If all that I'm experimenting on today in 2D will turn out well in 3D too, we'll have a whole new direction in implicit modeling.
Enjoying the work on MyTraL - sovereign athlete / personal training log: https://mytral.fitness/
I went on sabbatical to fulfill my dream project - consolidating years of training logs. I'm enjoying the technical challenges involved - digitizing paper hand written logs / visual models, navigating the maze of athletic metrics with their crazy trademarked names and multidimensional models. Having fun building AI coaches: agents ranging in character from Al Pacino in Any Given Sunday to the coach from my teenage years, utilizing ICL model-based predictions, ... and more.
The best part is the rush of memories while ingesting my own history - photos and recordings I completely forgot, as well as navigating data shared by friends.
This month has Strava & LeCol everesting challenge - signed up and added support that suits my needs to MyTraL. Good times.
I continue working on https://wireplug.org: A simple, free, and open source connectivity coordinator for WireGuard. Basically a way to keep WireGuard tunnels connected while moving between different access points. It handles (basic) NAT traversal and works with the in-kernel WireGuard driver on Linux and OpenBSD. You can find the technical details at https://wireplug.org
Input: video of an exercise/workout/dance. Output: single collage image as summary.
Me+team have been working on a large scope problem this past year. We're hardening some of our internal tools, and choose this small-scope problem about a month ago. Results have been encouraging so far, though not yet non-sloppy enough to share. We'll release in about a month.
Going into this, I thought I'd be spending most of my time tuning computer vision. Instead, the majority of the time has been handling codec and ffmpeg behaviour edge cases.
Working on gisti.ai which eventually will be a fleet of Product and Customer Operations AI agents that continuously analyze customer interactions, identify the churn risk and automatically route the right actions: code fixes to engineering, product insights to PMs, and operational tasks to customer operations.
Today, Gisti ingests customer feedback from every channel your customers already use, synthesizes it into a prioritized, evidence-backed list of opportunities, and lets product teams interact with an AI agent to explore, validate, and act on each one. We are building Gisti with a philosophy of complete automation for specific workflows.
We are looking for Design Partners, please hit me up at [email protected]. I will be in SF late August if you prefer in-person meetings.
Please ping me / email me if you are interested in a demo / demo account for deeper analysis.
I got so frustrated with all of the busywork that was building up in my email inbox that I built something better.
Housecat is the first email inbox that helps you get real work done. Our email connects to the other services you actually use like your Claude, Notion, Slack, CRM and Github and makes it easy to send messages and data between them. So if you need to update the CRM with a new lead, track a Github issue or ask a question of a colleague in Slack, you can do it all from right inside of your inbox. We just launched the email app in private beta.
Check out https://housecat.com/ or join the beta at https://home.housecat.com/welcome
Feedback and thoughts very welcome!
Making rent as an open source developer.
Desperately trying to attract new monthly sponsors and people willing to buy me the occasional pizza with my terrible HTML skills. Is it working?
Building http://zeroquarry.com
It's a way to augment small/overloaded security teams. It can pentest and then generate a pentester-style PDF report for auditors and procurement, triage incoming e-mails from security researchers by then checking whether the vulnerabilities they claim actually exist and are exploitable, hook into GitHub to scan for vulnerabilities and auto-propose fixes or file GitHub issues for you.
It's free for Open Source projects, if anyone here is maintaining one
I am working on building an self-improving QA agent for software teams. Free and open-source. It is an agentic testing harness with batteries, includes the test runner infra (web & mobile), memory (vector store, local embedding model powered by transformers.js, self improving loop, issue reporting. More details - https://vostride.ai | Code - https://github.com/vostride/agent-qa
Pulling together lots of solar data trying to promote solar panel installation in the UK.
We've had three back to back heatwaves so far this summer in the UK. THREE!
Hopefully it inspires people to make an informed decision to get the best price and starts generating more questions and leads for solar companies.
Two things right now:
1. A music recommendation tool based on Last.fm scrobble history.
It's graph-based, no ML or "fancy" techniques, but I've had good results with it so far. It builds recommendations based on a listening window or just recent history. It combines several different recommendation algorithms (including an Auralist-inspired "serendipity" score for novel recommendations, meant to simulate the serendipity of being recommended something novel by a friend), scores and ranks candidates, and takes in feedback that inform subsequent recommendations.
Fun project. Found some good new music with it already. :)
2. A code exploration / indexing tool with CLI and MCP interfaces for exploring concepts and impacts of changes in a codebase.
Essentially, an overwrought "find all uses" that doesn't depend on exact symbol or string matching.
I have a codebase of non-trivial size, but thankfully it's fairly well-structured. This tool indexes the code and bundles modules into "concepts" -- these can be auto-discovered or preconfigured. Dependencies, inheritance trees, symbols and symbol usage are all also indexed.
Then you can ask, "what's the impact of extending the domain model of XYZ" or "I want to remove this property" and it shows where to start, where to look next, and fuzzy edges or dependencies that might need deeper exploration. It surfaces non-obvious connections, too, or things a junior dev (or LLM) might miss, like when a model is mapped to an API DTO, or intermediary states, etc.
It's been useful for a new dev exploring the codebase, because you can ask in terms of business concepts instead of needing to know the exact symbol name in the code. And it's been much more token efficient than grep for exploration subagents. But it's limited to dotnet only.
I've been excited about OpenRouter's OAuth 2.0 PKCE flow for backendless BYOK inference. I just updated my Garmin watch AI app (https://untether.watch/en/apps/byok) to run the whole flow on the watch for a seamless BYOK experience - formerly you had to copy/paste your Gemini key into settings, which was a huge barrier.
If you're building client-side/frontend apps and want to let your users BYOK, OpenRouter's PKCE flow is great for that.
Otherwise I'm still working on Untether (https://untether.watch) - a suite of digital-minimalism apps that let you stay connected and do quick actions on the watch while keeping your smartphone out of sight (and your hands).
I am helping a not-for-profit, [NavSahyog](https://navsahyog.org/), build a custom software for their entire operations and programs that also tracks longitudinal impact. The organization conducts programs for children in Indian villages after school to build live skills.
This is my second iteration because the first version felt like a simplistic fit and improvement over their existing vendor provided app.
I have now designed a domain model based on my understanding and observations. I have a day job so I can't spend a lot of time in sync with the team. I have created a web app where the NGO management can test scenarios (by recording voice), and the AI (Claude Agent SDK) runs it past the domain model. In case, there is a gap, they can persist the scenario. After every iteration, I read through the scenarios and assimilate them into the domain model.
Several projects.
1. an extension to my shop (I have a shop! I never thought I would have one of these) for water treatment and food storage.
2. an agentic framework that I started on in January - maelstrom. You can find the current code here: https://github.com/zerohumancompany2/maelstrom-code/tree/ske...
The whole point of this framework is to increase the robustness of agents so that small models (30b class) can function over long time horizons reliably. Other goals include auditability (full agent sessions are stored durably and can be branched/rolled back/restarted from any point, all over-the-wire comms are also durably stored) and reliability (sane fallbacks for common failure cases).
The current iteration (sketch 7 or draft 7 or 8) is specifically a coding agent framework. In the future I'd like to expand the core to handle a variety of tasks.
Im building an agent that interacts with health care providers on your behalf to proactively get itemized bills and other substantiation as needed. Born out of my own frustration with the paperwork for getting reimbursed from my healthcare cost sharing ministry. The app is React with a Laravel backend on EC2 where the agent runs.
Recently Ive been experimenting more with coding from my phone using Claude Code for the Web. Its basically turned Github Actions into my development environment. Its enabled me to fire off a quick prompt in planning mode, go play with my kids, review and approve the plan while Im cooking dinner, then let it go 10 rounds with the AI code reviewer while I put the kids to bed. As a busy parent I feel way more productive than if I had to carve out sit down focus time for a side project.
Working on leaf: a terminal markdown previewer with GUI-like experience
Part of my job is to approve / reject MCP servers based on how secure they are and whether they are suitable for use in an enterprise environment. I was tired of my team being called the bottleneck to AI adoption, so I set out to automate the whole process.
I periodically collect the MCP servers and every new version from the Official MCP registry and assign them a score based on 29 distinct criteria like runtime guardrails (e.g. destructive tools, over broad permissions, rug pulls), SAST scans and transport & trust model.
As a result of this exercise, I found that 1 in every 10 MCP servers is pretty much unusable (score 40/100 or below). 18% of the popular MCP servers with 1000+ GitHub stars contain one or more security issues. 184 servers to date have changed their tool definitions after publication, which may indicate a "rug pull" attack.
I built this for security minded people who also want to be at the forefront of AI adoption and for security teams who are tired to be called the bottleneck.
Browsing the index is completely free, you only have to request an API key if you want automated, programmatic lookups for any workflow.
Feedback is always welcome!
A Remote Desktop tool for Mac and Windows https://forge.emrul.dev/agents/portal-desktop/releases
Still a few weeks away from getting everything working but it’s functional already and I have a half dozen regular users who’re my beta testers :)
In my free time, I'm building an iPad game to help Autistic kids practice real world skills in a virtual town - https://pocketown.app/
I've just released in beta, a web-based application called Verse Draft: https://versedraft.com
I wanted to create an all-in-one writing studio where fiction writers can keep all the details for an entire universe in one place while crafting stories, novels, movie scripts, TV series, or stage plays.
I also wanted the ability to allow for the limited use of AI in a way that only functions as a sounding board and does not write for the user; Where fiction writers could have access to tools such as a virtual assistant that they can converse with about their stories and world-building, but without it writing anything for them.
There is also an option to use the application without any AI tools at all.
Continuing to work on https://attachedapp.com.
ChatGPT validates your spirals. We help you prevent them.
We're in the middle of v2 over the next few months based on everything we've learned since launching ~a year ago.
As a habitual side project guy for the longest time, it's so satisfying to finally stick to one thing and go really deep. This thread has been highly motivating.
Still working on Study Engine and Nomnominees(more or less done for now).
StudyEngine is a webapp I'm using while doing my masters in comp sci. I upload lecture notes, textbooks, papers, etc. It then extracts topics and tracks my mastery of them over time. It uses an LLM to generate questions and flash cards. It loops in some newer learning science ideas. It tests recognition first(multiple choice), and then once a level of mastery is matched, it switches to recall. Working on adding RAG to it, so I can surface where in the source material something can be reviewed when going over quiz results. Currently just for me an some friends. If can get a good eval set up, I might work on optimizing cost and seeing if it could be opened up.
NomNominees is simple webapp that tracks James Beard, Great American Beer Festival, Festival of Barrel Aged Beers, and other awards. I use it when I'm traveling to find places to check out. Even just a cluster on a map shows me neighborhoods I might want to check out.
Recently worked on https://fastsleep.app - an app for nighttime overthinkers.
Instead of music or long podcasts, you are given something to imagine. Like if you hear "moonlight on a white flower", imagine that scenario until you hear the next one.
In short → Close your eyes, listen & imagine.
I just started running sword combat playtests this weekend for my next game, Today I Will Destroy You, which is about a girl who goes looking for her sister.
Combat is mostly inspired by Sekiro. Here's a minute of gameplay: https://www.youtube.com/watch?v=m8NJhd3Ks3k.
Already learned that it gets way too hard too soon, so this week I'm adjusting the training and difficulty curve to avoid frustrating new players.
If you liked Sekiro, have a PC, and want to playtest, please get in touch!
A daily meditation "instrument". I'm a big fan of Waking Up but I've kinda outgrown the catalogue. I know what to do now. I just need a timer and a couple of prompts...
I've been prototyping a bunch of old ideas with the free tier of ChatGPT, bouncing off design ideas and treating it like a pair programmer. Planning a blog post writeup of the general experience (along with saved chats in nicely cleaned-up Markdown) along with a handful of Show HNs.
I am working on hammer https://thehammer.io/ , it is a voice device that I have build and have been using for past several months for texting and getting answers to my questions via AI chatbots.
I started with the idea of replacing my phone with a texting device that can still keep me connected but realized phone has became utilitarian that it is not possible to replace it.
I still have to take my phone when I am outside but when I am home or at work, I now use hammer exclusively to text or to get answers. The most benefit I have got is that I don’t have the urge to open my phone and go on endless scrolling binge.
I'm working on a tool to automatically sync your work-in-progress music from your DAW (digital audio workstation, like Ableton Live, etc), to your phone. Basically a workflow tool for musicians and producers! Also has some quick share/feedback functionality too. Would love some feedback since it's still in really early stages. Already integrates with Ableton, Bitwig, Cubase, and GarageBand. https://trackid.com.
I used to work at Native Instruments, and super happy to now work on something for myself instead.
I spent the last 6 months building Elk Finder (https://elkfinder.com) and just launched a few weeks ago. It's a mapping web app similar to onX Hunt, except I have specific layers for finding ideal elk habitat in September and October (the primary archery and rifle hunting months).
Currently working on Agentry(https://agentry.run/) Runtime, connection, and deploy layer for building internal apps and automations with AI. You can build anything that you can build with Lovable, Replit etc. Your model, Your harness and your own Laptop/Server
some examples https://vessel-ops-dashboard-4c3976e79335c1aa.agentry.live/ https://local-goods-shop-54ce3b368e0d41f6.agentry.live/
I've just finished this chrome extension recently: https://ypuf.com/
It helps me to automatically save a tab that's not been used in a while so it auto-closes it but saves it as well as having the ability to snooze a tab like how you'd do it in gmail.
Everything is locally stored with 100% privacy in mind.
And vim like navigation is natively done.
I have been working on Stencilize: https://stencilize.me/
It's a little webapp that solved a problem I had when ordering PCBs: I was too cheap to buy the stencil when ordering the PCBs from China, but then I regretted it when I had to paste by hand. Because of this, I did the PCB Designer -> DXF -> CAD -> Add margin -> Add outline -> Print workflow by hand, but that became very tedious, so I built this to automate it.
It runs entirely on your local machine and it is hosted on Cloudflare pages, with the only costs for me being a domain name.
Building a small cloud service that I've been thinking of for a while, to take the articles I don't have time to read but find interesting in principle, and convert them to audio via TTS. Just filled a short flight with some articles from the backlog, now I need to do some UI improvements, and then maybe a similar flow for academic papers too. :)
Once my visiting family departs I am going to finish building the Little Free Library I started last month. We have lived in our current home for 5 years and it is about time I share my love of reading with my neighbors. https://littlefreelibrary.org/
Just trying to learn C again, making things from scratch in a multiplatform way, interfacing with X11 on Linux and wasm on the browser.
It's been fun dealing with memory and C's weird design in this age of agentic coding.
An operating system for self-hosting
https://github.com/malmoos/malmo
I saw the options out there were not fully open source, or had other limitations so I started working on this better one. Based on Debian, apps are docker containers.
I do work to adapt current open source apps, but it's so great to make them available as one click install.
I want to make it easy to run on a cloud VM or an old PC kept in the pantry. There are so many cases for self-hosting now that we need to make it easier to do
Automated network port change detection. Scan17 provides a solution to the question:
So your CTO decides to outsource firewall management - and the vendor carelessly leaves a network port open, exposing your production database. How does your team find out before an attacker?
Think of it as nmap port scan diff-ing. If a network port goes from closed to open you get an email or webhook alert. There is a REST API for automated workflows and privately hosted engines will be supported for some plans. There is a wait-list form on the website if you want to stay in the loop.
If you work in infosec / cyber security and are interested in being an early product designer / beta tester, let's chat! See my profile for how to get in touch.
https://digs.fm - like Goodreads, but for music.
As someone who's constantly on the look for new music to discover and being very deliberate about the things I'm listening, I needed a better way to organize the albums I want to listen to, listened and liked. And also I would like to see the discoveries of other folks who I know I like.
Original Show HN thread: https://news.ycombinator.com/item?id=32551862
I've been working on Source [1], an experimental Git server written in Go.
Source stores repositories in a database rather than a filesystem [2]. The primary goal is to rely on databases for durability, replication, and distribution, rather than introducing complex distributed filesystem infrastructure into the stack.
Building CVD Portal, a free whitelabel Coordinated Vulnerability Disclosure platform for EU manufacturers, plus tooling for the incoming Cyber Resilience Act. From September 2026 the CRA obliges manufacturers to file a 24 hour early-warning and 72 hour report to ENISA for actively exploited vulnerabilities.
Recent work is all top-of-funnel. Free no-signup self-assessment and exposure-scan tools that pull in people who don't yet know they have an obligation.
Live demo (no signup): https://demo.cvdportal.com