Launches
Every Show HN post, reviewed for whether people would use or pay for it. Strong is 6 or more, Solid 5 to 6; noise is Dismissed. Judged on the title and description only, so open the launch before you trust a score.
146 of 146 launches.
| Verdict | Launch | Category | Points | Age |
|---|---|---|---|---|
| 6.6Strong | Chiplab - Give your AI agent access to virtual hardware veecle.ai Real pain for embedded/firmware teams; virtual hardware for AI agents is novel if the simulation fidelity holds. |
DevTools | 7 | |
| 6.6Strong | go-git2 – 10-100x faster than go-git, libgit2 -> WASM -> Go github.com For Go devs hitting go-git's slowness and CLI divergence; read-only limits it, but the wasm2go trick is real. |
DevTools | 1 | |
| 6.5Strong | T0: We trained a 256M SOTA open-weights time-series forecasting model github.com Open-weights time-series model with HF weights; competes with TimesFM/Chronos but SOTA claims attract practitioners. |
AI infra | 8 | |
| 6.5Strong | Pippin – macOS High-Performance Screen Sharing Client for Linux github.com Solves a real pain for Linux users remoting into Macs without server installs; concrete and technical. |
DevTools | 2 | |
| 6.5Strong | The first open-weight model for general deterministic task huggingface.co Open-weight specialist-routing model for OCR/STT with confidence scores; credible if quality holds up. |
AI infra | 2 | |
| 6.3Strong | Every open government tender in the world (Over 250k) tenderportal.io Aggregating 250k global tenders in English has clear value for exporters; incumbents exist but data scope impresses. |
Data | 6 | |
| 6.3Strong | Faster, Cheaper Databricks Spark with Real-Time ML Config Management espresso.ai FinOps teams burning money on Databricks; real costly pain, but the pitch is thin on proof and rivals exist. |
Data | 2 | |
| 6.3Strong | Commit-Aware Router, route LLM traffic around your AI/cloud commitments github.com Enterprise FinOps with prepaid AI commitments; genuinely fresh routing angle, but it is an early LiteLLM PoC. |
AI infra | 2 | |
| 6.3Strong | Check your Claude Code, Codex and Cursor sessions for secrets github.com Secret-scanning for agent sessions addresses a real leak risk as agents handle credentials. |
Security | 1 | |
| 6.2Strong | Geneva, a Rust FFmpeg CLI alternative with HTML/CSS graphics genevarender.com Video devs wanting HTML/CSS overlays get a real ffmpeg alternative, but adoption hinges on edge-case parity. |
DevTools | 6 | |
| 6.2Strong | An Email-to-iMessage Gateway Email-to-iMessage gateway dodges 10DLC pain for notifications; clever hack but fragile and ToS-risky. |
B2B SaaS | 2 | |
| 6.2Strong | Muster, a Mac Menu Bar Switcher for Claude Code, Codex and Cursor github.com Menu-bar attention switcher for parallel coding agents; solves a real annoyance for heavy agent users. |
DevTools | 1 | |
| 6.1Strong | Geometry Check, a Fusion 360 add-in that finds slightly-wrong geometry github.com Fusion 360 users genuinely lose hours to near-square lines and gaps; a free detect-first add-in earns adoption. |
Other | 2 | |
| 6.1Strong | Merlin, Run Your Claude Code and Codex from Slack, Linear and GitHub usemerlin.ai Runs your own Claude/Codex agents from Slack/Linear on team machines; timely, but crowded with agent-frontends. |
DevTools | 1 | |
| 6.0Strong | Open-source Mac menu bar app for Claude, Codex and Cursor agentmuster.app Solves a real daily annoyance for people juggling multiple coding agents; free and open source helps. |
DevTools | 3 | |
| 6.0Strong | Graft – open-source source control for agent-scale monorepos github.com Agent-scale monorepo source control is a real emerging need; implementing Anthropic's design is timely but early. |
DevTools | 2 | |
| 5.9Solid | tsv has fast formatters and parsers for Svelte, TypeScript/JS, CSS github.com Faster-than-oxfmt formatters for web languages; credible benchmarks, competing in an active toolchain race. |
DevTools | 2 | |
| 5.8Solid | Jevcrawl – A web crawler for AI agents with no LLM in the crawl loop github.com Cheap deterministic crawling for agent pipelines; claims need verification but the pain is real. |
AI infra | 17 | |
| 5.8Solid | Tern – Native macOS SSH Client in Rust/GPUI (76 MB vs. Tabby's 580 MB) github.com Mac SSH users tired of bloated Electron clients; nice vault/sync angle, but Tabby/Termius/Ghostty crowd the space. |
DevTools | 2 | |
| 5.8Solid | Give Your Agents GPUs supercomputer.sh Teams whose coding agents need bursty GPU compute; per-second agent-managed sessions fit an emerging workflow. |
AI infra | 2 | |
| 5.8Solid | Ampsu – Fast calorie tracking with EU barcodes and adaptive metabolism ampsu.eu For EU calorie counters frustrated by US-centric barcode databases; solid niche with a real adaptive-target angle. |
Consumer | 2 | |
| 5.8Solid | Ace – Quizzes from your notes, every answer quoted from the source acenow.app Quiz generation from notes is a crowded edtech space, but source-quoted answers are a decent trust hook for students. |
Education | 1 | |
| 5.8Solid | Rhythm – a skimmable TUI for your GitHub repos github.com libgit2-via-WASM gives Go devs fast, CGO-free git reads; concrete benchmarks make it credible for tooling. |
DevTools | 1 | |
| 5.8Solid | Mac equivalent of Nvidia ShadowPlay. Recording missed moments on Mac clipit.itzadetunji.com ShadowPlay-style clip capture for Mac gamers; clear gap, working iOS version adds credibility. |
Consumer | 1 | |
| 5.8Solid | I aggregated 1,600 remote AI training jobs from 24 companies aitrainerjobs.co People hunting AI-training gig work get a useful aggregated job board with real data. |
Other | 1 | |
| 5.6Solid | Pulse Local, a Mac budgeting app whose AI models run on your machine pulseos.finance Privacy-first budgeting for Mac users wary of cloud finance apps; niche but genuinely differentiated. |
Fintech | 109 | |
| 5.6Solid | Deslop – Oxlint rules for AI-generated TypeScript github.com Useful lint rules for teams drowning in AI-generated TypeScript, but adoption depends on beating existing lint setups. |
DevTools | 31 | |
| 5.6Solid | AI Hub – Open Access Lesson Plans for Teaching Students about AI aihub.remakelearning.org Teacher-vetted library of AI lesson plans is genuinely useful for middle/high school educators, though free and niche. |
Education | 6 | |
| 5.6Solid | Whiteboard Defense – defend decisions of any system you've shipped github.com For devs shipping agent-written code who must explain it; useful angle, value depends on briefing quality. |
DevTools | 4 | |
| 5.6Solid | Sugabots – Group chats where your team and AI agents work together sugabots.ai Shared team+agent chat with real internal use; enters a very crowded space of agent collaboration tools. |
B2B SaaS | 3 | |
| 5.6Solid | Mac Markdown reader for your AI agents Markdown reader giving agent output a home outside terminal scrollback; real itch, free Mac app. |
DevTools | 2 | |
| 5.6Solid | Ruleherd – audits agent instruction files for contradictions and drift github.com For teams whose agent config files have sprawled; contradiction/drift checking is a real, growing pain. |
DevTools | 1 | |
| 5.6Solid | Agent-trials – a smoke-test runner for coding agent CLIs github.com For teams comparing coding-agent CLIs before adopting one; a concrete, timely eval harness in a real gap. |
DevTools | 1 | |
| 5.6Solid | Problems – see where agents get stuck using your product problems.dev Error-reporting channel for agents using your product; early but a genuinely new angle as agent traffic grows. |
DevTools | 1 | |
| 5.6Dismissed | macOS windows-like window maximize maximus.ethosoft.net Clone Windows-style maximize on macOS is a known itch already scratched by Rectangle and others; cheap price helps. |
Consumer | 1 | |
| 5.5Solid | GenClass, A 10 MB in-browser model that catches bugs that never throw npmjs.com In-browser model catching silent UI bugs in AI-generated storefronts is an interesting niche, but description is muddled. |
DevTools | 3 | |
| 5.5Solid | Fresh cleanup and reset a project after coding agent prs github.com One-command cleanup of agent worktrees and branches hits a real annoyance for heavy coding-agent users; still a small tool. |
DevTools | 3 | |
| 5.5Solid | BitRun – Run 1-bit LLMs on CPU at 98T/s without compiling github.com Tinkerers wanting to run BitNet models on CPU without build pain; genuinely niche but few such tools exist. |
AI infra | 2 | |
| 5.5Solid | Herdr for Android – The runtime coding agents run on droidtty.com Mobile companion for monitoring coding agents fits a real emerging workflow, but tied to an unproven 'Herdr' platform. |
DevTools | 1 | |
| 5.4Dismissed | Blobz: generate playful, deterministic SVG avatars from any string blobz.lirena.in Clone Another identicon-style avatar lib; fine React packaging but a crowded space with tiny stakes. |
DevTools | 7 | |
| 5.4Solid | FandomMap – a map of 700 fan conventions, with their guest lists fandommap.com Convention-goers planning trips around guests; aggregating 700 cons on one map is a useful, underserved niche. |
Consumer | 3 | |
| 5.4Solid | Mnemify – 3D terrain map merging Confluence, Notion, Obsidian github.com Unified 3D view of scattered notes is a novel twist, but terrain metaphors rarely beat plain search. |
B2B SaaS | 3 | |
| 5.4Solid | Free remote jobs API – 630k live listings across 16 job boards, no auth remote-jobs-api.tten.no Aggregating job boards is well-trodden, but free, no-auth, normalized access will attract job-tool builders. |
Data | 1 | |
| 5.3Solid | AgentGate – Fast authorization gateway for AI agent tools github.com Authorization gateway for agent tool calls addresses a real emerging need for teams deploying autonomous agents. |
AI infra | 2 | |
| 5.3Solid | Cut the tokens your coding agent spends reading tool output github.com Heavy coding-agent users watching token bills; real recurring pain, but compression tricks are easy to replicate. |
DevTools | 2 | |
| 5.3Solid | Agent Identity – Identity and communication infrastructure for AI agent agent-identity.dev Identity/comms plumbing for agents is a hot, crowded space; needs proof it's simpler than assembling pieces. |
AI infra | 1 | |
| 5.3Solid | Rewind Alternative citesilo.com Rewind-style local search of files and screen text had proven demand; a working cross-platform beta helps. |
Consumer | 1 | |
| 5.2Solid | Parsimony – an open-source benchmark for how much code AI models add parsimony.lasu.dev Measuring AI code bloat is a fresh, concrete angle on evals; adoption depends on the numbers being trusted. |
AI infra | 72 | |
| 5.2Dismissed | Fuel Gauge – Claude knows its own usage limits and stops at 95% github.com Thin wrapper Heavy Claude users hit limits mid-task; a 95% cutoff is a mildly useful unofficial workaround. |
DevTools | 4 | |
| 5.2Solid | Hatoba, an open-source SSH client with an AI assistant and E2EE sync github.com Sysadmins wanting an AI copilot in their SSH terminal; shipped and concrete, but SSH clients are a crowded graveyard. |
DevTools | 3 |