1v1 coding game that LLMs struggle with
Details
- External ID
- 47271751
- Source
- HN
- Company
- —
- Product
- 1v1 coding game that LLMs struggle with
- Website domain
- yare.io
- Launched
- March 6, 2026
- Cohort
- —
- Upvotes
- 29
- Upvotes percentile
- 0.7902829028290282
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
This is a game I wish I had as a kid learning programming. The concept of it is fairly similar to other coding games like Screeps, but instead of a complex world with intricate mechanics, Yare is a lot more minimal and approachable with quick 1v1 <3 min matches.It's purely a passion project with no monetization aspirations. And it's open source: https://github.com/riesvile/yareThe first version 'launched' several years ago and I got some good feedback here: https://news.ycombinator.com/item?id=27365961 that I iterated on.The latest overhaul is a result of simplifying everything while still keeping the skill ceiling high. And at least the LLMs seem to struggle with this challenge for now (I run a small tournament between major models - results and details here: https://yare.io/ai-arenaI'd love to hear your thoughts
Enrichment
- Theme
- Vertical
- Horizontal
- Function
- Dev tools
- Audience
- Developer
- AI stance
- Not AI
- Project type
- Hobby / open-source project
- Normalized one-liner
- coding game challenging for llms
- Manually corrected
- False
Could you build this?
Partial A real-time multiplayer code execution game requires sandboxed code execution, synchronized tick-based simulation, and anti-abuse safeguards that go beyond simple prompts.
What it would actually take: The architecture needs a deterministic tick-based game loop engine (Node.js/Rust) paired with secure isolated sandboxes (such as isolated-vm or WebAssembly runtimes) to safely execute arbitrary user-submitted scripts on every tick. It also requires WebSocket matchmaking, state synchronization, and cheat detection. The hard part is running dozens of concurrent sandboxed user code instances deterministically with microsecond timeouts without starving the server.
Discussion
8 comments analyzed.
Competitors mentioned: CodinGame - similar battle-based programming challenges
Concerns raised: LLMs lack feedback loop for iterative improvement, Unequal playing field - LLMs can't step through/simulate games like humans can, LLMs poor at spatial reasoning for this challenge, Powerful models (Gemini 3.1) failing suggests edge cases or strategy limitations, Marginal improvements from end-of-game reports vs. zero-shot prompting
Feature requests: Step through / simulation harness for LLMs to debug code, Better testing environment with feedback loop, Code review capability for opposing team submissions, Step(n), RunTil(), Board(n), BoardAscii(n), Log(m,n) debugging functions, ASCII board representation for game state inspection
Competitors
Other products that read as similar to this one — 90 launches clear the similarity bar, closest 8 shown.
Attention rank: #31 of 91 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 128 days after the earliest competitor.
- I taught LLMs to play Magic: The Gathering against each other · hn · 2026-02-17 · 117 upvotes · similarity 0.46
- Antimatter · hn · 2026-03-24 · 12 upvotes · similarity 0.44
- Shoot-em-up Deck Building game · hn · 2026-01-06 · 5 upvotes · similarity 0.43
- A real-time strategy game that AI agents can play · hn · 2026-02-25 · 220 upvotes · similarity 0.43
- TicTacToe · github · 2026-09-18 · 20 upvotes · similarity 0.40
- llmgame.ai · hn · 2026-01-06 · 28 upvotes · similarity 0.39
- Wolfrominoes · hn · 2025-11-23 · 10 upvotes · similarity 0.39
- Typocalypse · hn · 2026-07-04 · 5 upvotes · similarity 0.39
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a dev tools tool for Sales yet.