alt Hacker NewsAsk HN: What Are You Working On? (March 2026)
What are you working on? Any new ideas that you're thinking about?
Comments
Value-study-practice-o-matic is the best name I've come up so far. It's just an automated way of practicing rendering when painting.
A no-code platform for building SaaS apps with AI (serverless): https://saasufy.com/
Open-source security framework https://www.tirreno.com
https://eggbot.app/ an app for the widely OpenSource available EggBots to draw easter eggs. :)
Making improvements on this tetris meets block puzzle game
a general-purpose mobile agent, trying out on SLMs that work on phones. In case there's no internet.
I'm trying CM108AH with external ADC (for lower mic noise). I'm at the stage that I opened datasheets for both and started kicad.
I am working on Voiden : Api client based on executable markdown !
Check it out here : https://github.com/VoidenHQ/voiden
We are building a live knowledge graph of all political players in the South Asian Region. Essentially mapping out entities, relationships, and events with data from the last 30 years or so.
PayByFace is a nominee in the Romania Startup Awards 2026, we have about 1 day and 18 hours left to get as many votes as possible! Help us win this award if you believe in our project :) Vote here: https://strawpoll.com/05ZdzP64Qn6
mock, an API creation and testing utility. Any feedback is welcome.
https://rizful.com/ -- instant payments, anywhere, anytime, uncensorable, via the Lightning Network....
Hi! My name is Pablo. I’m a Product and UX Designer currently working on Maxxmod [1], a browser extension that gives users more control over the YouTube interface by reducing clutter, removing distractions, and adding features the platform doesn’t offer.
I’ve already completed the research, business model, competitive analysis, feature set, branding, and the full UI (40+ screens).
The MVP/V1 is currently in development. When the V1 is ready I’m planning to do a Show HN with this account.
It's my first product. Any feedback or questions are very welcome, even if it's just based on the idea and the screenshots on the site, since the product isn’t available to try yet.
Developing the AArch64 code generator for the DMD D compiler.
https://github.com/computerex/dlgo
Golang inference engine from scratch that can run a bunch of models with vulkan acceleration.
Developing this idea of a ClaudeVM and that being the future where we just write literate programs of Englishscript that run directly on the VM and eliminate this code compilation steps entirely.
Rebuilding iNaturalist with open source/weights models that can run on edge compute for better privacy, and hopefully accuracy.
Continuing my weekly newsletter about agentic coding updates:
Experimenting with operational transforms with https://rootdoc.app
I've been working on an online catan alternative. Play at https://sokataan.io I'm using expo and spacetimedb.
Nocode transform: From 30 photos --> FPS video game.
Example : https://shorturl.at/We3dH
Open Source screener of polymarket for insider tracking https://poly gains.com
A docker/container registry that deduplicates at the file level instead of the layer level. Faster pushes, cheaper storage costs.
The best agent framework: https://github.com/fugue-labs/gollem
i've been grinding away to make "Plates" the best it can be. i know building a weightlifting app is a massive cliche, but i felt compelled to make it anyway and honestly have zero regrets. i was fully expecting people to roll their eyes at yet another workout tracker, but to my surprise, the feedback has been incredibly positive.
even with everyone and their mother shipping a workout app (accelerated by the vibecoding surge), nothing quite hit the bar i was looking for. everything seemed to be built in react native, flutter, or another cross-platform framework, meaning it never truly felt at home on the device.
beyond just wanting something fully native, i felt there was a real gap in the prosumer market. i wanted to build something for people who already know what they're doing in the gym, have their program, and just want a solid tool they can actually own (it's a one-time purchase of $12.99) that at the same time gives them all the insights they'd reasonably want (muscle heatmap, 1rm breakdowns with different formulas, working volume, best volume, etc.).
the app is called Plates: Weightlifting Log. if you're interested in checking it out: https://useplates.com
Still working bringing AI agents to Godot. We recently hit 1k MRR.
Product link: https://ziva.sh/
Nobody will read this and the game is not ready, BUT maybe one of you can check it out and tell me what you think. It is called ZAEL. It is a mage arena web app (smartphone/Desktop), played from 1 to 4 players, no login, easy to share a link to invite people into your game, whatever the platform. It is no install, no bullshit. Just click and play
I have found out that it is very efficient to use phaser.js/three.js for fast, vibe coding, because it handles everything without having to setup a unity scene manually or unreal blueprint. I really recommend to make web apps instead for vibe coding. I love how easy it goes.
A prompt injection solution that seems to benchmark better than any other approach out there, while not using hard-coded filters or a lightweight LLM which adds latency.
Automated personal outreach app for job seekers, integrated with Gmail.
https://codedorian.com a programming language
Just launched DriftE — it's a in-depth Cloud discovery platform that fixes unmanaged "ghost" resources, configuration differences and manual cloud changes for enforcing your IaC.
I'm learning bug bounty hunting again. learn to contribute to NUCLEI open source contribution.
Recently, a simple desktop app for ingesting opentelemetry to help with other projects. This way, I dont need to set up another service just for that.
Still working on Bitwit, my CS/math spaced-repetition education site [1]
I'm dogfooding it heavily. The bugs at this point tend to be in card formatting. I ended up delaying the introduction of TeX/MathJax until I had quite a few cards written, and man oh man, it's tedious to go back and fix that formatting in hundreds of cards.
The real question: does it work? I _think_ so. I'm learning, and I feel like I'm retaining more, and I think the general structure counteracts or compensates for most of my issues as a learner. I think the science is pretty solid, but I'm also experimenting in a few areas, so... eh, we'll see.
If anyone is interested in trying it out, you can use it for quite some time (several months, maybe a few years) without needing to provide an email address or sign up.
A developer tool that lets you (or your coding agent) understand how users will experience your AI product before you ship it.
I'm building postaxis.io - it helps builders to distribute their own apps/products
Ishikawa : a framework/architecture for automated Attack Surface Mapping & Vulnerability scanning
- golang based architecture
- information is dynamically mapped into one central directed knowledge graph
- default multithreading
- utilizes existing tools (such as nmap/nuclei/katana/wfuzz/....) instead of reinventing the wheel
- architecture is (tldr) a self supervising logic in which every worker is also a scheduler that based on delta causality uses cartesian fanout and graph overlay mapping including local only witness nodes to dispatch new "jobs" without having a central scheduler or the necessity to scan a central total job queue to prevent duplicate executions.
In this architecture every "action" that can be executed defines an input structure necessary. If the previously mentioned mechanic identifies a possible job execution it will create a job input payload which will automatically be picked up by a worker an executed. Therefor every action is a self containing logic. This results in a organically growing knowledge graph without defining a full execution flow. It is very easy to extend.
I worked on this for the past ~10 years (private time). The sad truth tho is, while this project was initially planned to be open sourced - after i not to long ago for quite some bucks consulted a lawyer, i basically was presented with the fact that if i would publish it i could get sued due to germany's hacker and software reliability laws. So for now its only trapped on my disk and maybe will never see daylight.
Im right now working on a blog article (thats why i even mention it) about the whole thing with quite more detailed description and will also contain some example visual data. Maybe will post it on hackernews will see.
PS:The tool does not need llm/nn.
I'm currently researching how I transfer my risk management skills from traditional finance into decentralized finance. I'd like to put decentralized finance on a sound footing by building risk management infrastructure for it. One thing that has become apparent to me is that I have some deficiencies in my computer science skills. As a result, I've enrolled in a Masters of Science in Computer Science from CU Boulder through Coursera to remedy those deficiencies.
Following up the comment i made last month, I'm a solo dev building a handful of apps across different niches.
- Plask ( https://plask.dev ) — Google Analytics (GA4) connected analytics dashboard for people who ship multiple products. I got tired of manually checking separate GA4 properties for all my apps and SaaS projects, and setting up individual MCP integrations for each felt like overkill when I just wanted a quick overview. So I built a single dashboard that connects all your GA4 properties, runs statistical anomaly detection, sends alerts when something breaks, and generates AI weekly digests. Free tier for 2 properties, Pro at $9/mo.
- Kvile ( https://kvile.app ) — A lightweight desktop HTTP client built with Rust + Tauri. Native .http file support (JetBrains/VS Code/Kulala compatible), Monaco editor, JS pre/post scripts, SQLite-backed history. Sub-second startup. MIT licensed, no cloud, your requests stay on your machine. Think Postman without the bloat and login walls.
- APIDrift ( https://apidrift.dev ) — Monitors changelogs for APIs, SDKs, and libraries you depend on so you don't get blindsided by upstream breaking changes. Scrapes docs, diffs changes, classifies severity with AI, and sends digest emails. Track your dependencies, get alerted when something breaks. Free tier covers 3 sources with weekly digests. Built with Next.js, Supabase, and Gemini Flash.
- Mockingjay ( https://apps.apple.com/app/id6758616261 ) — iOS app that records video and streams AES-256-GCM encrypted chunks to your Google Drive in real-time. By the time someone takes your phone, the footage is already safe in the cloud. Built for journalists, activists, and anyone who needs tamper-proof evidence. Features a duress PIN that wipes local keys while preserving cloud backups, and a fake sleep mode that makes the phone look powered off during recording.
- Stao ( https://stao.app ) — A simple sit/stand reminder for standing desk users. Runs in the system tray, tracks your streaks, zero setup. Available on macOS, Windows, Linux, iOS, and Android.
- MyVisualRoutine ( https://myvisualroutine.com ) — This one is personal. I have three kids, two with severe disabilities. Visual schedules (laminated cards, velcro boards) are a lifeline for non-verbal children, but they're a nightmare to manage and they don't leave the house. So I built an app that lets you create a full visual routine in about 20 seconds and take it anywhere. Choice boards, First/Then boards, day plans, 50+ preloaded activities, works fully offline. Free tier is genuinely usable. Available on iOS and Android.
- Linetris ( https://apps.apple.com/app/id6759858457 ), a daily puzzle game where you fill an 8x8 grid with Tetris-like pieces to clear lines. Think Wordle meets Tetris. Daily challenges, leaderboards, and competititve play against friends.
Art search for magic cards
A developer tool that lets you understand how users will experience your AI product before you ship it.
Trying to solve source control collaboration for agents across dev teams to preempt merge conflicts pre-commit
Building HEBBS — a memory engine for AI agents, written in Rust.
The problem: every agent framework bolts together a vector DB for recall, a KV store for state, maybe a graph DB for relationships, and then hopes the duct tape holds. You get one retrieval path (similarity search), no decay, no consolidation, and the agent forgets everything the moment context gets trimmed.
HEBBS replaces that stack with a single embedded binary (RocksDB underneath, ONNX for local embeddings). Nine operations in three groups: write (remember, revise, forget), read (recall, prime, subscribe), and consolidate (reflect, insights, policy). The interesting part is four recall strategies — similarity, temporal, causal, and analogical — instead of just "nearest vector."
Some technical decisions I'm happy with:
- No network calls on the hot path. Embeddings run locally via ONNX; LLM calls only happen in the background reflect pipeline.
- recall at 2ms p50 / 8ms p99 at 10M memories on a 2 vCPU instance.
- Append-only event model for memories — sync is conflict-free, and forget is itself a logged event (useful for GDPR).
- Lineage tracking: insights link back to source memories, revisions track predecessors.
SDKs for Python, TypeScript, and Rust. CLI with a REPL. gRPC + REST.
There's a reference demo — an AI sales agent that uses HEBBS for multi-session memory, objection handling recall, and background consolidation of conversation patterns.
Still early. The part I'm wrestling with now is tuning the reflect pipeline — figuring out when and how aggressively to consolidate episodic memories into semantic insights without losing useful detail. Curious if anyone working on agent memory has opinions on that tradeoff, or if you've found other approaches that work.
Building a privacy first subscription tracker. https://paylog.space
moongate :) https://github.com/moongate-community/moongatev2 !
Not being fired because of AI
build tng.sh. nobody wants it. will kill it :)
was a fun project to do some rust
working on a text game engine similar to Evennia: https://github.com/electroglyph/atheriz
Is identity preservation technically possible for LLMs, or only information transfer?