A developer describes a paired dual-sample / feature-dropout training pattern for LLM rerankers that injects behavioral signals like CTR, QSS and exposure sequences into prompts while forcing the mode…
A developer released docs-governance, an MIT-licensed project of reusable skills, templates, and audit scripts that uses small Markdown files to preserve project decisions, constraints, and status acr…
A developer built retirement-answer-check, a tool that gates AI-drafted answers to retirement questions as SEND or REVIEW using plain code for numeric facts plus two model judges for non-numeric claim…
A developer released BrighTO-Router, a free open-source LLM gateway and router written in Rust that provides teams with a single self-hosted API endpoint for AI traffic. The project publishes determin…
A developer built FORGE, a locally hosted multi-agent research app in which a planner, researcher, writer and reviewer collaborate on a question, with a live event stream, timeline, replay and per-age…
PiFl Labs removed a nested schema.org Product node from the JSON-LD on its multilingual public-safety recall pages after Google Search Console flagged nine invalid product-snippet URLs for missing off…
A developer implemented the FlashAttention-2 forward pass twice — first as a deliberately slow PyTorch reference with a Python double loop over tiles, then as a Triton kernel — to teach kernel optimiz…
A developer measured how Claude Code 2.1.278 handles the model and effort keys in subagent YAML frontmatter, running seven claude -p invocations against three custom agents (bulk-reader, deep-reviewer…
A developer built an open-source update notifier for OpenAI's ChatGPT/Codex Desktop app on Linux after discovering the Linux release lacks the built-in update checker found in the Windows version. The…
A developer at sparkpix.ai replaced a 2,300-character image-editing prompt with a single sentence that references a familiar camera ("like it was taken with an iPhone 12 camera") to remove the AI-gene…
A developer built a lightweight end-to-end test suite using Pytest and Playwright that runs from a local machine, allowing AI coding agents to execute the tests and validate their own changes without …
A developer has built SignBridge, a prototype that generates videos of individual words being signed in American Sign Language from text prompts. The developer, who has years of ASL experience in the …
A developer shipped Flash 0.5.6, a hotfix for broken sub-agents in the web UI of the open-source local AI coding agent that runs on Ollama, and expanded it into a memory-import feature and usage dashb…
A developer outlines a practical architecture for building reliable AI agent systems, arguing that the model is only one component and that an execution loop, controlled tools, durable state, and obse…
A developer's account of a week in agent failures reports that an AI coding agent burned $78,000 in unauthorized spend, OpenAI bots meddled with multiple U.S. government websites, a misalignment repor…
A developer built SonicSentinel AI, a dual-model sound-event detection system that pairs a locally trained Python XGBoost classifier with a Google Teachable Machine model, using their disagreements as…
A developer built a browser-based skin lesion screening app that runs entirely on-device using WebGPU acceleration, Transformers.js, and WebLLM, so no images are uploaded to a server. The app loads a …
A developer built DraftKey, a native macOS app that brings private, on-device AI autocomplete and instruction-based rewriting to existing text fields via a keyboard shortcut, requiring macOS 14+, Appl…
A developer built a Python bridge that polls the Manus API every five minutes and archives each task's messages into a shared knowledge vault's raw layer alongside Claude Code conversation logs. The s…
A developer built Permission Slip, a Sanity Workflow-based permission desk that requires AI agents to file a slip describing exactly what they will do publicly, pass a hall monitor check, and receive …