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

I'm building a web-based local multiplayer party game platform. It's like a lovechild of Jackbox Games and Mario Party: https://gamingcouch.com. Back in December Gaming Couch hit the front page of Hacker News, you can check it out here: https://news.ycombinator.com/item?id=46344573

The TL;DR:

- Currently in free Early Access with 18 competitive mini-games.

- Players use their mobile phones as controllers (you can use game pads as well!)

- Everything is completely web-based, no downloads or installs are necessary to play

- All games support up to 8 players at a time and are action based, with quick ~one minute rounds to keep a good pace. This means there are no language based trivia or asynchronous games!

- In the future the plan is to open up the platform for 3rd party developers (and Gamejams!) as well. We'd take care of the network connectivity, controllers etc.. 3rd party devs can focus on developing cool multiplayer mini-games without spending an eternity with networking code and building the infrastructure.

I’m working on https://markway.io/, a Markdown web highlighter.

The core idea is to make it easy to capture page content, highlights and images and everything will be formatted as Markdown and is easily editable in a performant little editor window.

Then from there, I’m working on multiple export services like Clipboard export and popular PKM apps like Obsidian, Roam Research, Capacities etc.

https://codeinput.com - Tools for PR-Git workflows

Currently experimenting with semantic diffs for the merge conflicts editor: https://codeinput.com/products/merge-conflicts/demo

You can try by installing the GitHub App which will detect PRs who have a merge conflict and create a workspace for them.

I built RespCode because I was frustrated with the "one model, one answer" approach to AI coding assistants. When you're building production systems, blindly trusting a single AI's output is risky. What if it hallucinated? What if there's a better approach?

RespCode puts you back in control with 3 orchestration modes:

Compete Mode — Send your prompt to Claude, GPT-4o, DeepSeek, and Gemini simultaneously. See all 4 solutions side-by-side, compare approaches, and pick the winner. You're the judge.

Collaborate Mode — Chain models together in a refinement pipeline. DeepSeek drafts → Claude refines → GPT-4o polishes. Each model improves on the last, with full visibility into every stage.

Consensus Mode — All models generate independently, then Claude synthesizes the best parts into a merged solution. Democratic code generation.

But here's what makes it actually useful: Every piece of generated code runs instantly in real sandboxes — not simulated, not mocked. We support x86_64, ARM64, RISC-V, and ARM32 architectures. You see compilation output, runtime results, and exit codes in seconds.

Why "Human in the Loop" matters: AI models are powerful but imperfect. RespCode doesn't hide this — it exposes it. When you see GPT-4o produce clean code while Gemini's version has a bug, you learn which models excel at what. When Collaborate mode shows how Claude fixed DeepSeek's edge case, you understand the refinement process.

You're not just accepting AI output. You're supervising, comparing, and deciding. This is how I believe AI development tools should work — transparent, multi-perspective, and always keeping humans in the decision seat.

Here a detailed blog on why this could solve your coding problem - https://respcode.com/blog/why-single-model-ai-assistants-hol... Would love your feedback!

https://fooqux.com/ - an experimental article aggregator about software development. For several years now, I've had a routine of collecting articles on topics that interest me throughout the week and then reading them over the weekend. To help organize and streamline this process, I created this website. The main idea is to gather articles in one place and process them with a LLM — categorize them, generate summaries, and try experimental features like annotations, questions, etc. I hope this service might be useful to others as well.