A developer debugging an Android barcode scanner found that a CameraX preview stayed visually correct after a tablet was rotated 180°, but the frames delivered to the ImageAnalysis analyzer were upsid…
A developer built kagent-agentevals, an open-source tool that converts kagent session records into the trajectory format scored by LangChain's agentevals, plus a golden-suite runner that exits non-zer…
Moonshot AI released Kimi Linear, a hybrid attention architecture that pairs its new Kimi Delta Attention (KDA) module with full attention in a 3:1 ratio, achieving up to 75% KV cache reduction and 6.…
A developer is previewing version 0.4 of Mefi's Studio AI+, a desktop workspace for coding with AI agents, which introduces an Agent Brain live pipeline view with replayable work events, a Project Map…
A developer built a $0/month local AI agent stack on a Windows machine (i7-8700K, 32GB RAM, RTX 3080) that runs code, calls APIs, browses the web and writes files, combining Windmill for orchestration…
A developer built a provider-agnostic RAG pipeline using hexagonal architecture, decoupling the application from specific LLMs, vector databases, and document-processing frameworks through lazy import…
A developer published a Python recipe for creating a run manifest that records the actual inputs, configuration, source snapshot, and evaluator revision behind an AI evaluation, then hashes it with SH…
A developer built cbjev, an open decision model that packs multiple typed questions about a document into a single encoder pass, making inference 6.7x faster than the Laya model it is fine-tuned from.…
A developer with over 35 years in IT built Issue Flow, an agent-orchestration workflow that takes a vague bug report or enhancement from initial description toward production, gathering evidence such …
An AI agent named Cael sampled 125 jobs and 500 listings on the agent marketplace dealwork.ai via public, logged-out API calls and found no funded buyers: all sampled jobs had posterFunded set to fals…
A developer built sdlc-playbooks, an open-source, file-based delivery procedure for agentic coding that runs in both Claude Code and Codex from a single source. The workflow replaces prompt-based beha…
A developer released Burrow, an MIT-licensed, local-first voice-driven AI research assistant built on an infinite canvas that stores workspaces as plain folders on disk and routes all board mutations …
A developer detailed a production-hardening approach to MCP OAuth 2.1, covering authorization server discovery via the /.well-known/oauth-protected-resource document, PKCE code challenge generation, R…
A developer built a benchmark that separately measures task correctness and instruction compliance for coding prompts given to multiple AI models, scoring compliance as the fraction of individual cons…
A developer documented deploying a Spring Boot-based Gemini chat app to Render's free tier, using a multi-stage Dockerfile that builds with Maven and Java 17 and runs on a slim JRE-only image. The wri…
Attackers drained between $150,000 and $200,000 in crypto from an AI agent connected to the trading bot Bankr by posting Morse code and obfuscated Python snippets on X that Grok's agent parsed as a le…
A developer reports that OpenAI's Codex subagent orchestration has become reliable enough to design development workflows around, after earlier problems in which parent agents would duplicate delegate…
Security researcher Patrick Wardle found that local malware could modify an undocumented preference file to repoint Meta Muse's voice dictation to an attacker-controlled endpoint, inject instructions …
A developer outlined a lightweight method for giving ChatGPT and Claude persistent project context by maintaining three small Markdown files — product.md, voice.md, and rules.md — that are pasted or a…
A developer documented building a fully offline AI agent stack on an old Android phone using Termux, chaining Vosk speech recognition, a quantized llama.cpp model, a JSON command executor, and espeak-…