What are you working on? Any new ideas that you're thinking about?
I've been working on an LLM "harness" called Logbook[0] for fun with Codex.
The core idea was that I've always been a lousy notetaker, even going back to my school days years ago. I'm great at one-off and one-liner notes and occasionally more in-depth notes, but tend to not flesh them out fully enough to make them worth re-visiting.
This has been a struggle even as an engineer sitting in meetings or trying to absorb new information when starting a new job and ramping up.
Logbook is meant to use an interaction paradigm we as engineers are using very often these days: it's a terminal UI in the vein of Claude Code, OpenAI Codex, etc.
It's targeted at the entry of free-flowing thoughts but you can also write longer notes by launching your default shell editor from within the tool.
Each note is saved as markdown with some metadata and that metadata is then saved to a local SQLite DB.
For the LLM side, the tool extracts useful metadata from those notes and then performs some local ranking/categorization. It then has the ability to send a note or some metadata to a provider of your choosing (it's straightforward to use OpenAI or something more broad and customizable like OpenRouter) for further enrichment or filtering.
A couple examples of the currently implemented slash-commands: `/related` can be used to find related notes; say you've been scribbling down notes about OAuth or MCP servers and want to gather up the most relevant notes to one of those topics. Or you can use a `/gaps` command that'll help you find things you've taken notes about but without properly defining or providing context around them (i.e. you mention ID-JAG for OAuth but never actually say what ID-JAG is, this command will tell you this so you have a chance to review what you previously wrote and can then define exactly what that keyword is about).
It's still very much a work in progress. It's not meant to be a full-fledged note-taking app a la Obsidian or anything like that. I've just always preferred taking notes in markdown or plain text and this is a great way to continue doing that while also making enrichment of the notes pretty simple.
You may ask "why not just use agent memories?" I don't really like the idea of tightly coupling notes with codebases or agents and I don't find the current UX very intuitive at least for the way I prefer to take notes.
Repairing switching power supplies for IFR-1200S service monitors with my friend who's been in the repair business since the 1950s.
third-parties accounts for agents: https://sente.run/
agents have their own email and phone number and get a logged in browser instances on demand.
An agentic link discussion site: https://news.nuts.services
MakeSpell: autonomous crossword puzzle and more
Trying to make it effortless to build p2p apps without any setup:
A project made with jank and also my dotfiles.
secure and hide your files in plain sight.
MakeSpell: autonomous crossword and more
Working on Gaming Couch, a web-based local multiplayer party game platform. It's like a lovechild of Jackbox and Mario Party: https://gamingcouch.com
Just before the weekend I shipped a new mini-game called Pop It: Desert Island (https://gamingcouch.com/blog/pop-it-desert-island-launch). Launch went well: ~3,800 players from 56 countries over the weekend, and it immediately became the most played game on the platform.
It's a battle royale with an ocean/beach themed world, taking inspiration from Roblox, Mario Kart and others. The whole game is built in JavaScript (three.js for the 3D world) using a JS SDK I've been working on. It doubled as a test drive of the same SDK I want to launch for third-party developers, so anyone can build and ship a simple, fun multiplayer party game for the platform, ideally in a single weekend.
If you're a game dev, or aspiring to be one, and want to develop and ship your own party game check out this page https://gamingcouch.com/developers
The TL;DR of Gaming Couch:
- Free Early Access with +20 competitive mini-games.
- Players use their phones as controllers (gamepads work too).
- Completely web-based, no downloads or installs needed.
- Every game supports up to 8 players and is action-based, with quick ~1 minute rounds to keep a good pace. No language-based trivia or asynchronous (turn based) games.
Still plugging away on Raygum. Think Letterboxd for music.
I’m working on Ovio (https://ovio.au), a record-keeping app for Australian freelancers/sole traders/small businesses. The current version is deliberately narrow: send receipts by email/whatsapp; import bank transactions from CSV; and Ovio extracts the details and auto reconciles. From there you can check a BAS/GST-style summary and export organised records for an accountant.
It's been quite fun building this as this solves my exact problem, but trying to find an audience for a product is a completely different game
Working on a SaaS application that helps academic researchers conduct systematic literature reviews
In my 20 something experience of software development, it is totally ok if you don't work on anything so I don't work on something. If there is a possibility that your work will be something useful plus you will benefit from that, you definitely have enough time to do that in couple of coffee tea drinking times. Europe show off by their sidewalks and street signs. Computer is a little too lux for a human.
an ai first virtual computer as an mac app with mcp support
think ai is yolo sudo admin within the sandboxed linux
and internally/external you can control the whole computer via mcp, too
i'm working on a mobile app to control coding agents (claude code / codex / opencode / crush) from your phone
it uses https://sprites.dev/ sandboxes to run agents
I make wooden laser-cut custom maps. Right now I am working on making larger 2’x10’ versions
maroatlas.com
Working on a desktop application for llm evaluations.
I am working on a tinymush equivalent server in C#. I'm nearly at version 4 compliance.
still working on my local-first, byok personal finance tool. feels like a lot of sanding at this point.
Recently got a typewriter and I'm practicing on "long form" writing. Basically an offline blog.
I continue to take photos with film, developing and scanning at home.
I have been working on a starter template for Astro: https://github.com/BryanHogan/astro-starter-template
I've found Astro to be an amazing framework for simple, performant websites. It stays really close to basic HTML and CSS while adding useful features such as scoped components, layouts, and easy Markdown blog integration.
So I have been using it to build websites. But many things keep repeating with every website I build, so I began working on this project to create a base that I can use for every new web project.
It references content from my Clean Web Development Guide: http://webdev.bryanhogan.com/
When it is far enough along, I will use it for the landing page of the app I'm working on: a customizable solution for self-tracking including habits, health and journaling, or whatever else you need: https://dailyselftrack.com/
After more than 400 days of traveling around Korea, Macau, Mainland China, Japan and Australia, I'm now returning to Germany / Europe looking for work. I wrote about that in my monthly mail-letter: https://bryanhogan.com/follow
Working on a platform to create agents using prebuilt tools. Using it to learn more
RV64 toy/hobby kernel. No compatibility aim but rather at efficiency and speed.
I'm building Voxoria (https://voxoria.ai), it tracks whether B2B brands get mentioned when people ask ChatGPT/Perplexity/Gemini instead of Googling.
Ask the same engine the same question twice and you get different answers, different citations, sometimes a different opinion of your brand, so figuring out how best to present this has been a fun product problem to solve.
It also tries not to be yet another dashboard: instead of just analytics, an agent turns the findings into a ranked list of "ship this fix" todo items.
- a GUI (in python) for my "ancient" 3D printer to draw circuits on copper plates
- the gcode scripts are almost done !!
- a "customizable" mobile app (Android) for my business- a yet another static site generator (yaml, jinja2)
- a microcontroller for a hardware project (arduino)
- enhancements and reports for a desktop application (python)
Trying to rebuild the brakes on my Impreza. It is not going well.
So I am building https://buyvds.net with a global visual interface which has a list of around I think 249 vps providers over 60 countries combining up to 863 links (one vps provider can provide vps in multiple countries)
It uses DuckDB to expose a sql query interface in the website itself because I wanted to give the freedom to just do something interesting with the data.
My friend John had an idea which I really liked so I added "john mode" which shows what he was suggesting :-D
I think that Hackernews might like it but honestly, I have probably just made it out for myself and also as something to just share casually with folks on hackernews and other websites and hopefully I am able to help people and myself in some way with this website.
Open to some feedback as usual (for mostly all my projects really) and thanks for reading and have a good day dear reader and hey perhaps give my website a try!
tirreno security framework
A nondescript transcript-based collaborative audio editor.
A new model for assisted memorization, based on asynchronous interactions (using iPhone notifications): https://banyanflashcards.com
imagina.xplaya.com a site for my wife's stationary store in México. Customers ask for organizing images inside a printed page, I create a PDF for that
I’ve been building a shared memory layer across all AI tools
www.memoryplugin.com
A... database? for apps on a pocket switched network.
Currently hacking away at https://github.com/lockboot
Using UEFI SecureBoot + vTPM for cloud root-of-trust, a stack to prove what's released on github/gitlab is what's actually running on GCP/EC2 (and soon Azure & AliYun).
I was annoyed that so many companies in the Web3 space would do the on-chain theater of verified contracts and "audits" then 99% of their infra would be deployed on EC2 (or god forbid Vercel) in full un-ironic "Trust Me Bro" mode.
It's a different trust model from SGX/TDX, more pragmatic and hopefully easier/cheaper. Currently polishing off "Docker to verifiable cloud VM" stuff, and then gVisor support next.
I am building an newsletter for Short Science Fiction, using news stories of the day as inspiration. Purposely limiting the length to 3 chapters.
Currently removing the paywall from all my stories. If I have missed one, let me know.
The best damned ebook reader you'll ever use. Ready anywhere, ask anything.
We have been building https://merrilin.ai for 6 months now. Open to public access right now.
I'm working on Hiring Method (https://hiring-method.com). After ~2 years of development and two exhausting pivots, v1 is finally live.
I see a lot of new (and, to be frank, a lot of mature ones) HR tools are just wrapping Chatgpt around resumes (almost like "OK, now match this resume against this job posting and tell me if applicant fits"), which introduces a massive bias/inference problem.
I decided to build the exact opposite – a deterministic, math-driven fitness engine. It extracts structured scorecards from both CVs and job requirements and mathematically matches them, so you can actually review the exact reasoning behind why a candidate scored a, say, 85%. This fitness value is specified at every interview step – as applicant goes through an interview process their scorecard is updated at all steps.
If anyone here builds in the HR space, I’d love your feedback.
I'm learning to break 4-rotor Enigma encrypts.
needed seating planner for my wedding, so created something that suits my needs
Vibecoded a simple YouTube summarizer: https://github.com/tacone/prosey
The code might be slop, but it saves me time extracting information from long-form videos that would require 20-60 minutes of my time otherwise.
Also I get to not feed the algorithm just for the sake of curiosity.
1.mcp-memory : Persistent memory, a knowledge graph, code intelligence, and semantic search for LLM agents. Repo : https://github.com/corporatepiyush/mcp-memory
2.Secure Data Structures, Algorithms, Allocators, Thread Pools and Document parsing in C17. Repo : https://github.com/corporatepiyush/secure-c-lib
3. a coding agent that is cheaper, faster, more predictable, and dramatically more capable out of the box — because 584 of its 606 tools never touch a model at all. Repo : https://github.com/corporatepiyush/yantra-coding-agent
Needed seating planner to organize my wedding, came up with something that suits my needs
This thread was so long I vibecoded a clustering similarity graph of it to navigate with voronoi territories, should be available for the next little bit.
Got another daughter last week!
There were a lot of complications post delivery, and I want to make some sort of interactive story about it. We'll see how it goes
(Everyone is safe and sound)
Music!
Hoping to put out a project by end of year
An open pricing engine and app based on Quantlib.
https://quantra.io/