Daily Launch Index: March 22, 2026.

On March 22, 2026, I recorded 96 new product launches. This report provides a breakdown of what was built, how it was named, and the technology used to launch it.

This data helps you understand current market trends and see what competitors or fellow builders are working on right now. Use this index to find inspiration or to study successful launch strategies.

03 AM (8 projects shared)

• A Dominance Theory of Reductive Optimization in Complex Systems
• Sift, a local-first CLI for failures, root causes and next steps
• S2LC – 100 LoRA adapters in 3.59ms, zero HBM writes
• Generative Media Skills
• Agent Control – agentic runtime guardrails
• Deploy model whose predictions most resemble the ensemble mean
• CI-debugger – Debug GitHub Actions locally with breakpoints
• I built an AI teammate that takes Jira tickets and turns them into PRs

05 AM (10 projects shared)

• AWS Kiro accounts suspended en masse – AWS flagging fraud detection issues
• ez-stack - Stacked PRs for agents
• Zsweep – Vim Motions Minesweeper with Svelte 5.0
• Enable classic confinement for the gh snap (2021)
• Jensenify-MCP: "Every engineer should spend $250k/yr on tokens" –solved
• Mnesis – Lossless context management Python library
• Skill Lab – CLI tool for testing and optimizing agent skills
• Babel.nvim
• Void Text – a fast, native macOS writing app (~15MB, keyboard-first)
• Show HN : Query on-prem databases and DynamoDB in plain English

06 AM (6 projects shared)

• Synaphe – A type-safe language for hybrid AI and quantum computing
• Mindwtr – Open-source, local-first GTD app (Tauri and React Native)
• Quantum mechanics simulation Python library for research and learning
• Yeet – Throw AI tasks at hardware and walk away (Nomad and OpenShell)
• Banish: A declarative framework for rule-based state machines in Rust
• Entroly – Compress codebase context for LLMs by 78% using Rust

07 AM (1 projects shared)

• We've had front end reactivity since 2018 – via Async Iterables

08 AM (3 projects shared)

• DECT Cordless Phone Decoder
• Dune3d: A parametric 3D CAD application
• CCGears – .claude folder manager, Venv's for Claude Code

09 AM (5 projects shared)

• Autoresearch-genealogy: Structured prompts for AI-assisted genealogy research
• CLI for validating, querying, and updating Markdown specs
• Polybar: A fast and easy-to-use status bar
• Brezn – Decentralized Local Communication
• 0.1 and 0.2 = 0.3, – a two-strand number type for Python

10 AM (1 projects shared)

• Looking for Contributors and Sponsors – SuggestPilot

11 AM (4 projects shared)

• Ext-Markdown-mirror – now supports Pages Router and better image handling
• Flash-Moe: Running a 397B Parameter Model on a Mac with 48GB RAM
• Open source infrastructure (built in Rust) for internal software and AI agents
• Careless Whisper – personal local speech to text

01 PM (5 projects shared)

• CFO-stack: double-entry accounting setup for codex/Claude, inspired by gstack
• Cppsp v1.5.2 OOP system –Derive and Extension
• BDV – A privilege-separated Unix daemon for blind document verification
• Context Use – turn your data exports into portable AI memories
• Haven self-hosted Discord alternative

02 PM (4 projects shared)

• Local Cursor-A local AI agent that runs on your machine using Ollama
• CrypTorch: PyTorch-based Auto-tuning Compiler for ML w/ Multi-party Computation
• Gather- Self hosted community calendar
• OpenAI Parameter Golf: Fit the Best Possible LLM into a 16MB Artifact

03 PM (4 projects shared)

• One-command wrapper for AutoResearchClaw's 23-stage paper generation pipeline
• I built an open-source MCP server that parses game save files
• Compiling C to a custom architecture: my experience
• Tinyspec – A spec for tinygrad (ML compiler)

04 PM (7 projects shared)

• We replaced traditional ORM migrations with a DAG and stopped breaking prod
• Mnemo – a universal local brain for projects, usable by multiple agents
• Infinite Agent Canvas (FOSS)
• A Bridge Between Claude Code and Codex Using Channels and MCP
• Aegis – self-hosted secrets broker for teams priced out of CyberArk
• FlowState Dev memory that persists across AI conversations (MCP+SQLite)
• MultiHead: Turn one GPU into a team of specialized AI agents (open source)

05 PM (8 projects shared)

• Brand Toolkit – Claude Code plugin for framework-driven brand building
• CECbot: A TV box botnet that grabs the remote and maps the house
• Meow.gs – SSH to Dev Env with Touch ID / Face ID (Code on iPad)
• New development in PRNG of wyhash: w1rand
• US Job Market Visualizer
• Threejs 3D wireframe stylizing tool – Generate infinite variations
• OpenFOIA – local-first FOIA toolkit with entity graphs
• Showtime: SPEEDClaude Code Only Ships 100/100 Lighthouse Scored Code

06 PM (4 projects shared)

• Pcc – a C compiler in Python, powerful enough to compile Lua 5.5.0
• Gh-PR-dashboard – A GitHub CLI extension for every PR that needs your attention
• A Markdown file that turns your AI agent into an autonomous researcher
• AI agent for reading fast and learning new language

07 PM (5 projects shared)

• I built a photonic AI chip for space with 860x less power, rad-hard to 106 krad
• buddy
• kx/c.java
• Auto Translate JSON Library – JSON and Google to Multi-Format and 8 Providers
• Watchtower

08 PM (6 projects shared)

• Calculator C Parser
• Multi-node training on Mac Minis (PyTorch)
• BugPin: Self-hosted, open-source visual bug reporting tool
• Git-surgeon – Git add -p for AI agents
• Local UI for managing parallel AI coding agents
• Openpokertools.com

09 PM (6 projects shared)

• Ask HN: Is it feasible to build an ESP32 emulator in JavaScript?
• File exclusion reliability of AI coding tools
• ClawMem
• New OllamaMQ Version v0.2.5
• MFOS: Enforcing a commit boundary for AI-assisted decisions
• GitHub – DrBradStanfield/Roadmap

10 PM (4 projects shared)

• Noteriv – Open-source Obsidian alternative with MCP server for AI
• Bootstrap-Consistency Regularization (BCR)
• Test cases took my AI router from 82% to 98% accuracy
• AI DevOps Actions: 9 GitHub Actions for CI/CD in AI-Native Repos

11 PM (5 projects shared)

• I built a 200-line library to stop AI agents from going rogue (capkit)
• Apply video compression on KV cache to 10,000x less error at Q4 quant
• Zprof – a comptime-configurable and minimal memory profiler for Zig allocators
• ForgeKV – Redis-compatible KV server in Rust that scales with cores
• Claude Code for Academics [pdf]