Testing 20 Workers Forces Local Postgres Max_connections Scaling
I've scaled app servers, I've scaled HAProxy, I've scaled Postgres replica dbs, I've scaled job servers Never until today have I had to scale my local max_connections for Postgres but once you run 20 simultaneous workers doing full tests, now I have had to do even that. https://t.co/u89ZzEwOdF
Vercel Auto‑scales Build Hardware for Optimal Cost
Vercel can now intelligently pick the right hardware for your build. With new Rust-based compilers like Turbopack & Rolldown, build performance now scales with 𝒪(cpus). But too many CPUs and you waste money. Too few and agents waste time. Elastic build machines...
Self‑Review Tool Boosts PR Accuracy 3‑8× Daily
seriously @walden_yan cooked, this thing legitimately saves my ass 3-8x a day, and yes it sounds weird that devin can catch devin's own mistakes, but this is basically the equivalent of "sleeping on it" and looking at a PR with...
AWS Now Adds IDs to Security Group Rules
I thought there was a problem with the security group rules created by my bootstrap script initially but there was not. AWS added ids to security group rules which threw me for a loop in my tired state when I...
Claude Cowork Switches to API, Needs Clearer UI Testing
I keep trying to test my UI using Claude Cowork and it will do a few UI tests and then started hitting the API directly. I need more explicit instructions.
Google Releases 60‑control Checklist and Terraform for Cloud Security
346: Zuckerberg Finally Finds His People, They Are All AI Agents One does not simply walk into cloud security - but Google just published a 60-control checklist and some Terraform to help you try. Ryan loves it, but what does...
Virtual Sandboxed Containers Become Essential for Emerging Agents
New agent primitives are sprouting up left and right. Virtual sandboxed containers for them are one of the hot new infra components…
Turn Bulky MCP Servers Into Lightweight Binaries for Agents
Maybe flip some heavy MCP servers to tiny, fast binaries that your agent can use within a Skill? That's what @iRomin did here as an experiment and I think the approach has merit. https://t.co/if8SC0djvj
Anthropic's Claude Code Gains Fully Autonomous Auto Mode
Holy moly, these guys never stop shipping. 🚨 @AnthropicAI just pushed a major update to Claude Code, moving closer to true autonomous execution. Meet 'Auto Mode' This drop effectively patches the primary bottleneck in CLI-based AI agents: the constant need for human-in-the-loop authorization...
Test Data Entry via API After Inspecting Network Traffic
using playwrite mcp for testing data entry. I'm really glad I write all my apps as clinet-server with an API. Its will do the first few in the UI, inspect the network traffic and then start posting directly to the...
Production AI Agents Require Full Observability Metrics
If you can't answer "who did what, when, why, and with what data,' your AI agents aren't ready for production. #DevOps #Observability https://t.co/tRGwCPc4Mb
Built AWS Batch Environment in 2.5 Weeks, Not Years
What I've Vibe Coded 🤖 In 2.5 Weeks ~ Compared to similar code I tried to implement for years to deploy an AWS environment for running batch jobs (as AI agents or not). How I did it. https://t.co/BlXxVvHagH https://t.co/btfZ0Yw1hK
Airweave AI Automates Root‑Cause Diagnosis From Production Logs
🚨 @Airweave_ai just solved one of the absolute hardest problems in production debugging and... ... they open-sourced it 🤯 Server logs are great at telling you when something broke in production. They are terrible at telling you why it happened. This new tool adds...
Map AI Dependency Risks Before Provider Outages
Most AI teams have a dependency problem they haven't mapped. If your compute provider had a two-week outage, what breaks in production? If you can't answer that clearly, that's your next architecture conversation. https://t.co/skfSBDmvnp
AI Speeds up PHP Tests From 20 Minutes to 3
My full PHP test suite took 20+ minutes. Unbearable. Yesterday, I asked Claude Code to speed up my test suite. I knew that parallel testing would be a massive time saver, but I had set it up so that things immediately...
AI Agents Need Product‑style Lifecycle Management, Not One‑offs
AI agents rarely fail where most teams expect them to. They don’t fail in development, where everything is controlled and tested. They fail later, once exposed to real-world variability. I see this repeatedly. An organisation builds an agent, the demos...
GStack Streamlines Massive Legacy Codebase Integration
Just created my first PR into the Y Combinator codebase (1.84M lines of code total, this PR was 2400 lines) I got the YC internal/bookface codebase to run natively in @conductor_build - and it was hairy since it's a gargantuan...
Master Fundamentals Before Diving Into Kubernetes
If you’re starting your DevOps journey, don’t begin directly with Kubernetes. Many beginners jump into advanced tools because they are popular and in demand. But without strong basics, things start feeling difficult very quickly. Before Kubernetes, get comfortable with: • Linux • Networking...
10 PR Bug Fixes and CI Test Refactor Improve Stability
Cooking with GStack today. I just dropped 10 PR bug fixes from the community plus a big refactor of E2E CI tests, which should help stability overall. https://t.co/xJDTIpixTy
Add Temporal Developer Skill to Claude Code
For those of you, developing with Temporal Temporal Developer Skill - install in Claude Code using https://github.com/temporalio/skill-temporal-developer
Reimagining Software Factories with AI: 8090 Leads
I think the concept of building a Software Factory is now a commonplace expectation. Yay. The winner still isn’t clear but whoever does the best job reimagining the software development lifecycle in a world of agents, AI, expert knowledge, tribal...
Claude Code Flags Dependabot PRs that May Break Cloudflare Workers
Claude Code giving useful reminders for GitHub workflow dependabot PR recommendations that may break my app's usage on Cloudflare Workers platform 🤓 👍
AI Agents Can Now Maintain the Software They Build
We're early in a world where agents maintain the software they help you build. Ramp Labs team is proving it works at production scale and the results are only getting better as models improve. Sharing what we learned, hope it's...
Metrics Alone Won’t Ensure Happy Developers—Prioritize Platform UX
You can have green DORA metrics and still have miserable developers. Speed ≠ Joy. To get real IDP adoption, we need to treat platforms like products. APIs aren't enough. Read my new article for Platform Engineering here: https://platformengineering.org/blog/the-empathy-gap-why-your-platform-needs-ux-not-just-apis #DevEx #PlatformEngineering #devopspink
Windows CI Now Fully Tests and Builds Libghostty‑vt
Big day for Windows + libghostty-vt: - We now run the full unit tests on Windows in CI - We now build libghostty-vt in CI - We fixed a ton of issues related to MSVC with libghostty-vt - CMake now supports building libghostty-vt...
AI‑generated Code Makes Testing Essential, Says Momentic
In this episode of Founder Firesides, YC's @harjtaggar talks to @wuweiweiwu and @itsjeffan, co-founders of @MomenticAI, who just raised a $15M Series A. Momentic is the verification layer for software. The AI-powered testing platform impersonates end users to catch bugs before...
AI Still Edits Migrations, Not Creates New Ones
still weird that after all this time, AI still tries to edit my migration files instead of just creating new ones. some database habits are apparently eternal.
Key Differences Between Agent‑Native and Cloud‑Native Infrastructure
We just snuck out this cool little @googlecloud paper that spells out what to think about when preparing agent-native infrastructure. What changes from a cloud-native approach? Direct link: https://t.co/10AiOuo7LG https://t.co/7Eg6IqHaJa
GStack: Accelerate Software Projects and Build Faster
I released GStack 2 weeks ago and we're still growing I made it because I was realizing: every software engineering project is changing. How could I accelerate the way I saw @steipete accelerate? GStack is my gift to everyone who wants...
Serverless End‑to‑End MLOps with FastAPI and Lambda
End-to-end ML deployment breakdown: • Train & serialize model • Build FastAPI inference layer • Containerize with Docker • Push to ECR • Deploy via Lambda + API Gateway Serverless. Scalable. Production-ready. This is real MLOps 🔥
Dual AI Workflow Doubles Quality of Specs and Reviews
Created a chatgpt 5.4 plugin for Claude, it automatically gets a "second opinion", forges the best results for prd/spec/implementation. Once finished and reviewed, submits to chatgpt for bug review / security review analysis. Works insanely better having two work together. ⏺...
GStack Relies on Codex for Every Code Review
GStack actually will use Codex for code and plan reviews and I use it 100% of the time now
Boost macOS AI Coding with Cmux
Yeah. If you’re coding with either Claude Code or Codex in a macOS environment, use cmux. https://cmux.com/
AI Codegen Hype Ignores Deployment, Observability, and Reliability
The chatter about generating code with AI tools feels stuck at the "basic" level of... well, codegen, plus (perhaps) reviews and testing. I hear close to little talk about the things that come right after generating code: deploying, canarying, o11y, SLOs,...
Kickstart Secure AI‑Powered Coding with Kiro CLI
All the complicated stuff I do with AI starts here. I still use this method with a few tweaks. Instead of writing a book I publish here. Learn to write good code with AI ~ and secure it. Implement A Simple...
Secure AI Factory: Cisco‑Nvidia Full‑Stack for Scalable, Private AI
What does a Secure AI Factory look like in an enterprise run by humans and AI agents? @Cisco's answer is Secure AI Factory, a full-stack architecture designed to generate tokens at scale while keeping data private and operations manageable. And it's...
GitHub Repo Cuts Claude Code Token Costs 70%
Found a Github repo that 10x your Claude Code. 25 agents. 108 skills. 57 slash commands. Token optimization baked in. Agents handle architecture, security review, TDD, refactoring, E2E testing. Each one scoped to a single job. Best part: token costs cut 60-70%. Sonnet default. Thinking...
Iterate in Small Steps: Safer Than Big Overhauls
One lesson from years of DevOps work: Small changes are safer. Large changes feel efficient but carry hidden risk. That idea seems just as true with AI assisted development. Iterate quickly, but in small steps.
Silencing AI Noise to Build Practical AWS Automation
I’ve started muting all the accounts pushing hot takes on ai 🤖 so I can focus on getting things done. If it’s not contributing towards accomplish something not really helping. Here’s what I’ve gotten done or trying to get done with...
AI Can Instantly Double Your Test Coverage
Okay this might be totally obvious but you can just tell Claude Code to double your test coverage and it will actually do it. Will need a lot of review, but if you’re just getting started with adding tests to your...
AI's Future Lies in Workflow Orchestration, Not Models
I am starting to believe that the most interesting part of AI development is not the models. It is the workflows around them. Agents coordinating work, orchestrating tasks, managing context. Feels a lot like the early days of DevOps when we started wiring...
Removing Depth Limit Boosts Agent Success From 58% to 88%
Spent two weeks benchmarking Pilot on Terminal Bench 2.0. Ran 500+ tasks across 15 experiments. Built analysis pipelines. Measured variance. Compared agent behavior across pass vs fail runs. The fix that moved the needle? Removing one env var that forced maximum thinking...
AI Simulates Six Months, Cuts Build Time 30%
In 48 hours our AI agent simulated 6 months of activity in the app. Anthropic dropped a 1 million context window last weekend and we put it to work. No human intervention. The results: • 30% faster build times • 50% reduction in...
GStack Adds Windows Support, Falls Back to Node.js
GStack just shipped Windows support. Thanks to all my Windows users for bearing with us. There was a Bun runtime error on Windows and now it falls back to Node.js. Big thanks to the GStack contributor community - sorry it...
Redesign Feedback Loops, Don’t Just Add AI
RT Your new backlog shouldn't "add AI." It's redesign feedback loops so models, pipelines, and platforms learn together. Think: model drift to incident response, feature flags to guardrails, SLOs to AI behavior. #AI #DevOps #Agile @Star_CIO https://t.co/7dcoLIKa0K
Prioritize Indexing, Caching, and Transport for Faster APIs
20 steps to improve API performance ⚡️ DB + code: indexing, query caching, pooling, efficient algorithms Caching: Redis/Memcached, HTTP cache headers, CDNs Transport: GZIP/Brotli, HTTP/2–3, keep-alive, TCP tuning Scale: pagination, async processing, load balancing Reliability: rate limiting, timeouts, proper errors Ops: monitoring/profiling, versioning UX: smaller payloads, better...
Agentic AI Empowers Solo Architects, yet Breeds Unextendable Black‑box Apps
Agentic AI can turn a single architect into a "full-stack squad" in minutes, but it can also create black-box apps that no one can safely extend. #AI #DevOps https://t.co/p18hdtdbZn
Kubernetes 1.36 Adds Native Scale‑to‑zero Pods
If only my Kubernetes pod could scale to zero. That'd be great for staging/test environments or irregular production workloads. Oh, that's coming in Kubernetes v1.36 after sitting in alpha for years? Sweet. https://t.co/klzNA6Hs0X https://t.co/Yee1mpTLEh
Turn Intent Into Tests, Free QA for Quality Thinking
This direction is very meaningful, since translating intent into executable tests can reduce friction in QA workflows and let teams focus more on quality thinking rather than repetitive implementation.
CLI Version Beats Prompts and Node Upgrades
Node 18 + ClaudeCode 2.1.72 is a cheat code 😉 We benchmark Pilot on Terminal Bench 2.0. 89 real coding tasks, Opus 4.6, Modal containers. Ran 10+ full experiments over two days. The CLI tool version matters more than prompt engineering, effort...