Anthropic announced five developer platform changes for the Claude API, including the removal of fast mode on Claude Opus 4.7 effective July 24, 2026, after which requests with speed: "fast" will retu…
A developer created pi-tiered-router, an extension that routes each phase of a coding agent to a different model, optimizing cost and performance. The system breaks a coding turn into classify, plan, …
Developer Dan Nwaneri built Nightwatch, a system that uses a Cloudflare Worker and an LLM to filter live soccer match events and send Telegram alerts only for high-significance moments, such as goals …
Google ADK's built-in LiteLLM integration allows developers to use models like Claude and GPT-4o instead of Gemini without changing agent architecture. A DevOps assistant built with Claude Haiku demon…
A developer built Slate-memory, a one-shot attractor memory system that enables the cheap Claude Haiku model to match the accuracy of the expensive Claude Opus model at 1/640th the cost. The system us…
A developer built fivedaylaunch, a website studio for local service businesses that delivers hand-designed sites in five business days for $25/month with no upfront fee. The tech stack uses Node.js, S…
A developer built an AI agent for NHS cancer waiting times that refuses to predict individual wait times, instead providing honest historical odds for each trust. The agent uses a data pipeline to pro…
AgentSafe Labs retracted a previous finding about Claude Haiku returning UNCERTAIN results in red-teaming tests, revealing the cause was a bug in its own safelabs-eval detector logic rather than ambig…
A developer building a personal knowledge system reduced Claude API costs by up to 84% by routing vision OCR tasks to the cheaper Haiku model and concept extraction to Sonnet. The technique separates …
Claude Fable 5 can be used as an orchestrator in multi-agent workflows, delegating execution to cheaper models like Opus or Sonnet, cutting token costs by 10x without quality loss. The approach involv…
Alan Turing Institute researchers discovered a workflow-level jailbreak in GitHub Copilot that bypasses safety guardrails by decomposing harmful requests into ordinary multi-turn coding tasks. Testing…
Anthropic researchers introduced the Jacobian lens (J-lens), a method that reveals a sparse workspace of verbalizable concepts in language models carrying multi-hop reasoning and hidden cognition, dis…
Researchers at barkup-bench ran seven pre-registered studies with over 13,000 model runs to determine the most reliable way for LLM agents to edit structured data. They found that giving every node a …
VetoBench, a new open-source benchmark for AI memory, tests whether agents re-propose approaches teams have already rejected. In trials, agents with no memory violated prior decisions 80-90% of the ti…
Developer Tommy Tran released Traceburn, a local-first tracer and efficiency profiler for AI agents, which identified 69% avoidable spend in a test run by suggesting prompt caching. The open-source to…
Claw Patrol launched a security firewall for AI agents that intercepts and gates outbound requests to databases, Kubernetes, and APIs. The tool parses traffic at the wire, enforces rules via LLM judge…
A developer explains how to distinguish between genuine API rate limits and network-level failures when using OpenAI, Claude, or Gemini APIs. The post details diagnostic steps using curl, DNS checks, …
A developer team reduced their LLM API costs by 60% through systematic optimization, starting with instrumenting every call to log token counts per request type. The biggest savings came from semantic…
Enki Labs released evaluation results for its closed-source memory engine Enki, showing comparable answer accuracy to mem0 on the LongMemEval-S benchmark while storing roughly half the facts (138 vs 2…
A developer released nine open-source Telegram bot templates on GitHub, covering use cases such as AI assistants, booking systems, crypto alerts, job channels, price monitors, restaurant reservations,…