I built an experimental system to test whether an autonomous AI can propose actions freely but be structurally prevented from executing side-effectful actions without explicit authorization. The system consistently blocks unsafe filesystem, shell, and network operations and produces a trace and diff proving nothing changed, even under adversarial pressure. The goal was to see if refusal and non-action can be enforced and verified at the execution layer rather than relying on prompts or logging.
Ask HN: What are you working on? (January 2026)
341–350 of 914 posts
Re: Ask HN: What are you working on? (January 2026)
#342I'm working on an AI (Aethas, https://blog.aethas.ai) that preps context before my meetings, sets reminders, makes design drafts, summarizes meetings, and drafts follow-up emails without me asking. Basically trying to have an entire personal AI like Jarvis that has the same context as me but acts on it automatically. Uses my Obsidian for knowledge, Claude for reasoning, pull data from written notes... Still early days but it feels like magic when I use it.
Re: Ask HN: What are you working on? (January 2026)
#343ALT (https://www.APILoadTest.com): I recently built ALT as a no-code platform to load/stress/performance test APIs with pre-defined (and customizable) load profiles, via globally distributed machines. The goal is simple: key performance metrics, in seconds, without scripting. I also added the feature to directly import collections from Postman.
Here's a walkthrough of how ALT works: https://youtu.be/m2MER0KW3yA
Re: Ask HN: What are you working on? (January 2026)
#344Nowhere near as impressive as some of the things here but I’ve been slowly growing out a daily puzzle website https://regularly.co - I built the first few games for my wife, no subscriptions, no pay to play, no sign up needed. Im launching a new word game next week which I’m super excited about. If you do play it and have any feedback do shoot me message!
Like the game. for no place king X the ux would be faster for me if i could draw a line where no kings can go.
Re: Ask HN: What are you working on? (January 2026)
#345Building a text editor w/ git-like version history
versionary.app
Re: Ask HN: What are you working on? (January 2026)
#346Nowhere near as impressive as some of the things here but I’ve been slowly growing out a daily puzzle website https://regularly.co - I built the first few games for my wife, no subscriptions, no pay to play, no sign up needed. Im launching a new word game next week which I’m super excited about. If you do play it and have any feedback do shoot me message!
Just tried Kingly, really liked it!
Re: Ask HN: What are you working on? (January 2026)
#347Also building a CW Ham radio from a kit
Re: Ask HN: What are you working on? (January 2026)
#348Spent about 2 years improving printing and scanning stack of Linux: CUPS, SANE, AirSane, as well as some legacy drivers, and also x86 proprietary driver emulation on ARM with Box86. Even that "modern" printing stack in Linux is 20+ years old, there's still such an unbelievable amount of basic bugs and low-hanging-fruit optimizations, that it's kinda sad. Not to mention that it still maintains ALL its legacy compatibi…
Thank you. I have thrown printers out the literal window.
He's our hero
Re: Ask HN: What are you working on? (January 2026)
#349Turtle graphics with lambda functions. I don't know if it has any interesting applications, but I wanted to learn lambda calculus
Re: Ask HN: What are you working on? (January 2026)
#350Spent about 2 years improving printing and scanning stack of Linux: CUPS, SANE, AirSane, as well as some legacy drivers, and also x86 proprietary driver emulation on ARM with Box86. Even that "modern" printing stack in Linux is 20+ years old, there's still such an unbelievable amount of basic bugs and low-hanging-fruit optimizations, that it's kinda sad. Not to mention that it still maintains ALL its legacy compatibi…
Great initiative. I wonder, how likely is it for a complete beginner to break their own printer or scanner by making a mistake in driver implementation? Or is it possible to work on hardware support without having a physical device? I assume it is impossible to test each and every one printer and scanner, so there is probably some clever tricks there, right?