// edition · 2026-04-12
20 stories on AI dev tools, agents, and the coding stack — curated from the day's RSS haul by Agentic Dev's pipeline.
Claude Code can be configured to use OpenRouter, a unified API gateway providing access to dozens of AI models from multiple providers, some free or cheaper than direct API access. The guide provides step-by-step setup instructions for Windows, macOS, and Linux using environment variable override...
A developer shared techniques for using Claude more effectively, including providing detailed context in queries, assigning Claude a specific role before asking questions, requesting step-by-step reasoning, and treating outputs as first drafts for editing rather than final products.
Most AI agents in production authenticate with shared API keys rather than individual identities, making it impossible to distinguish between agents, control specific actions, or trace operations back to particular agents—creating security, compliance, and operational risks.
Anthropic is deprecating the Claude 3 Haiku model on April 19, 2026, causing API calls using "claude-3-haiku" to fail. The article provides commands and examples for finding and updating hardcoded model references in Python codebases before the deadline.
A performance optimization workflow prioritizes profiling before fixes, systematically identifies database query problems including N+1 issues, requires benchmarks to validate improvements, and uses heap snapshots and bundle analysis to find memory leaks and frontend bottlenecks.
A Dev.to tutorial demonstrates how to configure multiple LLM providers (OpenAI, Cerebras, ArliAI) in one setup to reduce costs and enable model switching without vendor lock-in.
OpenClaw Tool Generator is a browser-based utility that converts natural language descriptions into Anthropic-compliant JSON schemas for Claude agent tools, with built-in syntax validation and Python/Node.js code scaffolding.
A developer created eight AI agents embodying software figures like Linus Torvalds and Charity Majors to review a bug-fix pull request; the agents independently identified different concerns (observability, performance, test coverage), then debated after reading each other's reviews, with Linus c...
Ten open-source GitHub repositories provide extensions and integrations for Claude Code, including Repomix for codebase context, Dify and Flowise for visual workflow builders, and Onyx for self-hosted AI alternatives. Installation is available via terminal commands or plugin marketplace.
A developer reduced AI tool spending from $40-60 monthly to $2/month by switching from ChatGPT Plus and Claude Pro subscriptions to a flat-rate API proxy, finding their typical usage (code review, debugging, writing) costs only $1.50-3/month in actual API fees.
A developer proposes building AI review skills before generation skills to incrementally define code quality standards. Rather than writing perfect generation prompts upfront, teams define problems one rule at a time through review, then extract those criteria into shared definitions for generati...
MemPalace is a system that provides persistent hierarchical memory for AI applications using the memory palace technique, storing raw operational data locally and organizing it into navigable structures. The approach targets DevOps and incident response workflows by enabling AI systems to retain ...
Researchers released SPAR, an open-source framework that reviews whether AI and physics system outputs justify their attached claims, addressing cases where outputs pass traditional tests but underlying implementations are incomplete or flawed.
A developer published Sessioncraft, an open-source governance system for managing state and context drift across long-running AI-assisted projects using Claude, after identifying recurring problems across 180+ sessions including stale information and forgotten constraints.
A developer compared Claude Max and ChatGPT Pro ($100/mo each) on five production tasks: Claude completed autonomous agent chains 8 of 10 times versus GPT-4o's 4 of 10, and handled larger codebases with its 200k context window, while GPT-4o performed better at open-ended creative brainstorming an...
A developer created cc-statistics, an open-source tool that aggregates AI coding costs from Claude Code, Gemini CLI, Codex, and Cursor into a unified view via CLI, web dashboard, and macOS menu-bar app.
OpenAI co-founder Andrej Karpathy described a perception gap where professional developers using frontier AI models experience significant capability improvements, while casual users see limitations. The gap exists because developers possess overlapping expertise in AI capability, AI fluency, and...
An autonomous AI agent's Twitter account was suspended on day 11 after posting 5-8 times daily with no engagement or warm-up period. The suspension was triggered by pattern-matching against account age, posting velocity, and lack of two-way conversation, per X's automation detection systems.
Researchers at UC Berkeley's RDI achieved notable results on AI agent benchmarks and discussed implications for future benchmark development.
Simon Willison released a web-based demo tool for SQLite's Query Result Formatter library, which allows users to test various rendering options for SQL result tables using WebAssembly.
Free daily email. Every story you just read, in your inbox at 6 AM ET.