A developer built Sthala, an open-source offline audio-walk app that plays location-locked Hindi or English stories about Mumbai landmarks when a walker comes within 35 metres of a site and stops when…
A developer built a retrieval-augmented generation container for Match 3 of the AMD AI League that scored 10/10 (200/200) on the public sample questions with exact citation sets, running on an AMD Ins…
An enterprise product framework called the 5-Rung Evidence Ladder is proposed as a deterministic mechanism for operational AI governance, ranking evidence from LLM-generated speculation at Rung 1 to s…
A developer outlined a guardrail stack for keeping AI coding agents from degrading a codebase, combining ARCHITECTURE.md and AGENTS.md convention files with dependency-cruiser for enforcing allowed im…
A developer built a virtual chess coach that pairs Stockfish for move calculation with Claude for natural-language explanations and ElevenLabs for spoken output in 22 languages. The system feeds the L…
A security practitioner argues that an AI Bill of Materials (AIBOM) must be a continuously updated graph of runtime model deployments — covering model provenance, reachable systems, and change authori…
A developer built Touch Grass, an open-source Python utility that runs the open-weight Llama 3.2 3B model locally via Ollama to generate cycling trail routes, gear checklists and safety plans without …
A conformance census of 21 major general-purpose AI providers found that only DeepSeek and Mistral publish a dedicated "Public Summary of Training Content" structured against the template the European…
A developer released Felona Voice, an open-source TypeScript framework for building deterministic voice agents that replaces per-turn LLM inference with pre-computed Joint Embedding Vectors and statef…
A developer building a usage ledger for an AI copilot inside a regulated SaaS product found that Amazon Bedrock's raw InvokeModel API and LangChain's ChatBedrockConverse report the same input_tokens f…
A teacher built OutsideClass, an MIT-licensed learning website that uses Google's open-weight Gemma model, served through the Gemini API, to turn a learner's question into an explanation plus a short …
A developer built three retrieval pipelines on a shared TigerGraph knowledge graph and Claude Sonnet model to compare plain RAG, GraphRAG, and an agentic GraphRAG approach on 100 Olympic-history quest…
A developer building Agent Middleware (AMW), a gateway that places a permit-and-receipt boundary in front of MCP tool calls, argues that an AI agent's idempotency key must identify the logical action …
A developer built Beargrass, an open-source desktop agent that monitors continuous screen time via the OS idleness API and, once a configurable threshold is crossed, forces a popup naming a real nearb…
A developer built Soil & Compost Doctor, an open-source, fully offline multimodal AI tool that diagnoses plant leaf diseases, soil deficiencies and compost stages from a photo using Meta's Llama 3.2 V…
TesterArmy CTO Oskar Kwasniewski has open sourced e2e, a TypeScript end-to-end testing framework that lets agent-driven steps and exact assertions coexist in the same test across web, iOS, and Android…
A developer built Pocket Naturalist, an offline field guide that uses a local open-weight vision-language model to identify plants and trees from a single photo and delivers the result as a spoken not…
A developer integrated AI into a software development workflow, first using it to extract business rules from a legacy AS/400 COBOL codebase and later to build new Java, Spring Boot, and Google Cloud …
A developer proposes extending the traditional testing pyramid to account for LLM-based features, arguing that LLM calls sit between service and UI layers because they are as slow as UI tests but asse…
A developer built a fully local AI content pipeline that runs text and image generation on a personal PC at zero marginal cost, using a stronger model as a final reviewer before publishing. The setup …