#chipotle burrito bowl 5,419. Thanks @chipotle.

CNN’s Harry Enten: “Congrats to the GOP-held Congress, you did it! Fewer Americans (10%) now approve of you than think we faked the moon landing (12%). The record high disapproval (86%) is driven by a huge shift from the GOP. Their net approval of Congress is down 89 pts from last year to -56 pts!”

#chipotle burrito bowl 5,418. Thanks @chipotletweets.bsky.social

#chipotle burrito bowl 5,417. Thanks @chipotletweets.bsky.social

#chipotle burrito bowl 5,416. Thanks @chipotletweets.bsky.social

#chipotle burrito bowl 5,415. Thanks @chipotle.

#chipotle burrito bowl 5,414. Thanks @chipotletweets.bsky.social

Claude Code Routines: Anthropic's Answer to Unattended Dev Automation - DevOps.com Anthropic's Claude Code Routines let dev teams automate scheduled tasks and API-triggered workflows from managed cloud infrastructure.

Anthropic made it easier to automate dev work. Claude Code Routines let you schedule tasks, trigger workflows via API, or react to GitHub events — all from managed cloud infrastructure. No cron jobs. No babysitting.
devops.com/claude-code-... #ClaudeCode #DevOps #AIAutomation

OpenAI Upgrades Its Agents SDK With Sandboxing and a New Model Harness - DevOps.com OpenAI's updated Agents SDK adds sandboxing and a model-native harness, giving enterprises a controlled way to build long-horizon AI agents.

OpenAI just updated its Agents SDK — and the additions are ones enterprise teams have been asking for. Sandboxing. A model-native harness. And a clear signal that the Assistants API is on its way out. Here's what changed and why it matters.
devops.com/openai-upgra... #AgenticAI #OpenAI #DevOps

Systems Thinking: Thriving in the Third Golden Age of Software Every few decades, someone declares software engineering dead or about to be automated away. It hasn't happened yet — and why it hasn't is worth understanding if you're a developer trying to figure ou...

The third golden age of software started around 2000 — not with AI. AI is just the latest level of abstraction. Grady Booch puts it in historical context in my final piece of The Agentic Shift series for #CoderLegion:
coderlegion.com/14459/system... #AgenticAI #SoftwareEngineering

#chipotle burrito bowl 5,413. Thanks @chipotle.

#chipotle burrito bowl 5,412. Thanks @chipotletweets.bsky.social

#chipotle burrito bowl 5,411. Thanks @chipotletweets.bsky.social

The Re-Soloing Risk: Preserving Craft in a Multi-Agent World There's a version of AI-assisted development that sounds like progress but functions more like regression. A developer sits alone, orchestrating ten agents, each working a different branch of a codeba...

AI generates code fast. Your CI pipeline doesn't. The bottleneck has moved from writing to validating. New piece for #CoderLegion on why testing is the new speed limit — and what to do about it:
coderlegion.com/14458/the-re... #AgenticAI #DevOps

Azure MCP Server 2.0 Brings Self-Hosted Agentic Automation to the Enterprise Microsoft releases Azure MCP Server 2.0 with self-hosted remote support, enterprise security hardening, and sovereign cloud readiness for agentic AI workflows.

Microsoft released Azure MCP Server 2.0 — and the headline feature is self-hosted remote deployment. Teams can now run MCP as a centralized internal service with enterprise security controls, sovereign cloud support, and 276 tools across 57 Azure services.
techstrong.ai/features/azu...

Claude Code Can Now Run Your Desktop - DevOps.com Anthropic's Claude Code now controls your Mac or Windows desktop. Here's what the computer-use feature does and the security risks.

Anthropic's Claude Code can now control your Mac or Windows desktop — open apps, navigate browsers, manage files — while you're away from your desk. Early stage, real implications.
Full article: devops.com/claude-code-... #ClaudeCode #AIAgents #AnthropicAI

The Validation Bottleneck: Why Testing Is the New Speed Limit AI coding agents can generate code in seconds. Your CI/CD pipeline cannot keep up. That gap is the next major problem in software development. And it's one the industry is only beginning to take serio...

The "plumbing" of agentic AI is getting commoditized. Sandboxing, state, orchestration — platforms are handling it. What remains hard: failure modes, observability, security. New piece for #CoderLegion:
coderlegion.com/14457/the-va... #AgenticAI #AI

#chipotle burrito bowl 5,410. Thanks @chipotletweets.bsky.social

Visual Studio Code 1.115 Moves Deeper Into Agent-Native Development - DevOps.com VS Code 1.115 adds the VS Code Agents companion app, better tools and background terminal interaction for agentic development workflows.

Visual Studio Code 1.115 is out — and it ships a brand-new companion app built for agent-native development. Parallel sessions, background terminal interaction, smarter browser tools. Here's what's new.
devops.com/visual-studi... #VSCode #AgenticAI #DevOps #DeveloperTools

GitHub Copilot CLI Gets a Second Opinion — and It's From a Different AI Family - DevOps.com GitHub's new Rubber Duck feature in Copilot CLI adds a second AI model from a different family to review code plans.

GitHub just added "Rubber Duck" to Copilot CLI — a second AI model from a different family that reviews your agent's work before it executes. Pairing Claude Sonnet with Rubber Duck (GPT-5.4) closed 74.7% of the performance gap vs. Claude Opus alone.
devops.com/github-copil... #GitHubCopilot #AICode

#chipotle burrito bowl 5,409. Thanks @chipotletweets.bsky.social

From Prompts to Goals: The Rise of Outcome-Driven Development Most AI coding tools work the same way. A developer spots a problem, writes a prompt, and the agent executes. It's useful. But the developer still drives every single decision. Google appears to be re...

Google's Project Jitro: tell the agent your goal (lower error rates, better test coverage), let it figure out the path. The shift from prompts to outcomes is bigger than it sounds. New post for @CoderLegion:
coderlegion.com/14455/from-p... #AgenticAI #DevTools

#chipotle burrito bowl 5,408. Thanks @chipotle tweets.bsky.social

#chipotle burrito bowl 5,407. Thanks @chipotletweets.bsky.social

Google's Scion Gives Developers a Smarter Way to Run AI Agents in Parallel - DevOps.com Google's open-source Scion testbed lets developers run isolated, parallel AI agents across local and remote clusters. Here's how it works.

Google open-sourced #Scion — an experimental testbed for running multiple AI agents in parallel, each in its own isolated container with its own credentials and workspace. It's early, but the approach is worth watching.
devops.com/googles-scio... #AIAgents #MultiAgent #OpenSource

Microsoft's new SQL MCP Server gives AI agents structured, secure access to enterprise SQL data — no exposed schema, no fragile NL2SQL. Just deterministic, role-aware queries built on the Data API builder. Open source and free.
techstrong.ai/features/mic... #MCP #AzureSQL #AgenticAI

#chipotle burrito bowl 5,406. Thanks @chipotletweets.bsky.social

Google's next coding agent isn't waiting for your prompt. Jitro, the internally referenced Jules V2, is reportedly designed around goal-setting — you define the outcome, the agent figures out the code changes.
devops.com/googles-next... #AIAgents #DevOps #GoogleJules

Anthropic just launched Claude Managed Agents — a full production stack for AI agent deployment.
No more months of infrastructure work before you ship. Secure sandboxing, state management, and multi-agent orchestration — all handled.
techstrong.ai/features/ant...

Microsoft Agent Framework 1.0 is here — production-ready, open source, built for enterprise multi-agent AI.
It replaces Semantic Kernel + AutoGen, supports OpenAI, Anthropic, Google Gemini + more, and gets you to a working agent in 5 lines of code.
techstrong.ai/features/mic...