Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: What are you working on? (August 2026)

news.ycombinator.com

Post reply on HN
Print view View original

861–870 of 1001 posts

Re: Ask HN: What are you working on? (August 2026)

#861

Post by jagmeetchawla »

After 30 years of corporate life, I'm rediscovering the magic of building - and yes, coding. Deep into Rust, Swift, and Claude Code these days.

One problem I keep hitting: with unattended sessions, or something done a while ago, I find it difficult to reconstruct what I asked Claude Code to do and what it actually ended up doing. Yes, there are sessions I can scroll, and there are jsonl files I can grep. And yes, there are session replays. But it is still quite time consuming. Not to mention going through all the git diffs. And then there is the issue of multiple sessions coordinating their own agents on the same repo.

So I'm working on a CLI tool that helps me reconstruct this interval by interval, and then do a deep dive where needed. So far it has been quite interesting - it is teaching me a lot about how the Claude Code harness works.

If you have a similar problem, I would be very interested to know how you solved it.

Re: Ask HN: What are you working on? (August 2026)

#864

Post by RichardChu »

I'm working on self-hosted local email infrastructure for agents and apps: https://github.com/churichard/fluxmail

It consists of a MCP server, a CLI, and a REST API. Everything is run and hosted locally on your computer. You can use the MCP server to give your AI agents access to email, use the CLI to interact with your email, or build your own email apps on top of the REST API.

The goal is to make it simple and easy to interact with your email through any interface you choose, and to give you control over the way that you use your email.

It supports Gmail, Outlook, and IMAP/SMTP. Give it a try and let me know what you think!

Re: Ask HN: What are you working on? (August 2026)

#866

Post by abrichr »

I’ve been working on OpenAdapt: https://github.com/OpenAdaptAI

It’s an open source attempt at making GUI automation reliable enough for things like EHRs, ERPs, Citrix apps, etc.

A lot of agent approaches basically keep looking at the screen and asking a model what to do next. We’ve gone in a different direction. We use models to learn or repair a workflow, but once it works we try to compile it down into something deterministic. On a normal run there can be zero model calls.

Lately I’ve been working on figuring out how to know that an action actually had the intended effect, rather than just that the click/keystroke happened. So workflows have preconditions, postconditions/evidence checks, and halt if the state is ambiguous. We also recently added a way to send those halted decisions to a phone for a human to approve/reject.

Re: Ask HN: What are you working on? (August 2026)

#867

Post by schmuhblaster »

I’ve been working on DeepClause [0], an agentic harness, the core of which is based on Prolog. Every interaction (conversation turn, plan, execute plan, etc.) triggers a small logic program that orchestrates one or more agent loops or prompts. All these programs are completely hackable, so you can easily adjust the core logic and behavior of your agents in a reliable manner. Also, instead of using markdown specs, you can “compile” your markdown into Prolog code. My hope is that this might unlock new possibilities for spec driven development.

It’s been a lot of fun and I am somewhat proud of Prolog/Typescript integration layer built on SWI Prolog’s WASM version.

Other than that, I am not sure how and if I will continue with it. Feedback welcome!

[0] https://github.com/deepclause/deepclause-sdk

Re: Ask HN: What are you working on? (August 2026)

#870

Post by wbnns »

I left my job at the end of June; I'm tired of playing someone else's music and want to write my own songs (build my own apps)

So I'm building an app a week and focusing on solving niche problems I've experienced in my own life to start

I think it's never been easier to build prototypes, but because of all of the slop, building community and finding users around them are harder than ever before

This latter part I think is going to be the hardest part of going out on my own

But going to try

Things I'm working on right now:

https://a2ciple.pt/ - testing prep for a language exam required for residency & citizenship in Portugal

https://basehub.org/ - a docs site for agents to get the latest context on Base, an Ethereum L2

https://playgokab.com/ - a vocab game to learn words in diff languages

https://mru.space/ - a protocol to run software that gracefully degrades across failing hardware

https://shiponeshot.com/ - a framework to help me ship prototypes more quickly

Anyhow, cheers :)

Post reply on HN