Sign in

Casey West

@caseywest.bsky.social
232 followers 32 following 53 posts

Climber, thru-hiker, acrobat, cyclist, dad, and leading AI Infrastructure & Applied Engineering for Developer Relations at Google Cloud. caseywest.com

PostsRepliesMedia
Casey West @caseywest.bsky.social · 12h
There's an endpoint that scores whether you can get the Spot VMs you want, 0 to 1. Most teams have never called it. But it's advice, not a reservation. I watched a zone fall 0.90 to 0.10 in 4 hours. Part 2 of 9: caseywest.com/obtainabilit...
caseywest.com
Obtainability, Part 2: The API That Tells You What You Can Get
There's a beta Compute Engine endpoint that answers "can I get twenty Spot VMs of a given machine type in this region right now?" with a number between zero and one, and most teams on Spot have never ...
000
Casey West @caseywest.bsky.social · 24/09/2026
Spot teaches you to survive losing a machine. It doesn't teach you what to do about the one you never get. My advisor said 0.90. Three regions gave me zero L4s for 36 hours. Part 1 of 9 on the half of scarcity nobody built for: caseywest.com/obtainabilit...
caseywest.com
Obtainability, Part 1: The Other Half of Scarcity
You went to Spot VMs for the discount and did the homework: checkpoint code, a drain handler, a preemption survived. Then a job sat in Pending for days across three regions and never got a single L4. ...
001
Casey West @caseywest.bsky.social · 27/08/2026
My AI-generated code percentage is embarrassingly high, and it's the wrong number to chase. It's this era's lines-of-code: pulled from a diff, easy to game, silent on whether anyone can explain the system in six months. caseywest.com/the-code-generation-percentage-your-org-shouldnt-be-measuring
caseywest.com
The Code-Generation Percentage Your Org Shouldn't Be Measuring
"Percent of code AI-generated" is this era's lines-of-code: easy to compute from a diff, trivially gamed, and silent on whether anyone can explain the system in six months. Here is what the field data...
000
Casey West @caseywest.bsky.social · 25/08/2026
Three Kubernetes projects added brakes to their own elasticity. Kueue froze a quota field. LeaderWorkerSet limited its autoscaler. Agent Sandbox slowed its refill loop. Same window, same kind of fix. I think I know why. caseywest.com/three-kubernetes-projects-shipped-a-brake-pedal-this-month
caseywest.com
Three Kubernetes Projects Shipped a Brake Pedal This Month. Why?
In one week of late-August 2026, Agent Sandbox, Kueue, and LeaderWorkerSet each shipped a limiter on their own elasticity: a warm pool that refills more slowly, a field that used to be editable and is...
100
Casey West @caseywest.bsky.social · 20/08/2026
Your OKF bundle passes validation. It can still be wrong: a new source contradicts a page from last year, and the spec never notices. A timestamp records a fact, it doesn't run a check. So I built the system to manage your LLM-Wiki. okfctl: caseywest.com/a-field-is-n...
caseywest.com
A Field Is Not a Process
A timestamp records when a wiki page was written. Nothing in the format runs to keep it true. Twenty-one years after I helped build a commercial wiki, the LLM-wiki idea is back, and it dropped the one...
120
Casey West @caseywest.bsky.social · 13/08/2026
When the accelerator you want is gone, don't just retry. Rank the ones you'd accept, fall to the next, and climb back when the best returns. That's a fallback ladder—and on GKE, one ComputeClass object. caseywest.com/compute-scarcity-is-p…
000
Casey West @caseywest.bsky.social · 11/08/2026
Idle AI agents don't burn cycles. They hoard RAM. Google's GKE Agent Sandbox benchmark: the isolation lever is worth 44%. Suspending idle agents and reclaiming their memory is worth better than 3x on its own. The sandbox is the smaller half. caseywest.com/your-agents-are-too-a…
000
Casey West @caseywest.bsky.social · 10/08/2026
I usually work with engineers. Last Wednesday I got to open the first stop of Build with Gemini, in NYC, with the business track. Finance, HR, ops. No code, and they still shipped an agent that waits on a human before spending money. cloud.google.com/events/build...
000
Casey West @caseywest.bsky.social · 07/08/2026
A Pending pod that "should" fit—free chips everywhere, none adjacent. I built that into a browser puzzle: real H100, A100, TPU slices, five levels, no signup. Running clusters like this and explaining them is the job I'm hiring for. caseywest.com/schedule-the...
caseywest.com
Schedule the Accelerators: A GPU/TPU Puzzle
I'm hiring a Senior Developer Relations Engineer for GKE and AI Infrastructure. Instead of describing the work, I built it: a small, playable cluster scheduler that teaches the real problem behind run...
000
Casey West @caseywest.bsky.social · 06/08/2026
A team asks for Claude + web search in prod, this quarter. The architect reads three cells—tier, toggle, contract—and says yes before lunch. A lookup, not a risk assessment. Series finale. New? Start at Part 1: caseywest.com/governed-gro...
caseywest.com
Governed Growth, Part 4: Multi-Model, Governed, and Boring
The finale of the series: the mature end-state the first three parts build toward. When every server-side capability's trust boundary is a decision on record—the deployment tier you chose, the toggle ...
000
Casey West @caseywest.bsky.social · 04/08/2026
No single off switch for "every server-side tool." 5 controls across 4 layers—partner web search denied by default. Part 3 is the runbook I've shipped: which toggle, at which scope, in gcloud + Terraform. Finale next. caseywest.com/governed-growth-part-…
000
Reposted by Casey West
Remigiusz Samborski @rsamborski.bsky.social · 03/08/2026
👀 Sneak peek behind the scenes of Google Agent Skills! In my new article, I share how we build, test, and scale AI skills at Google: • Standardized layout • Remote MCP tools for IAM • Automated CI/CD checks • 2x2 evaluation matrix Full article link in replies 👇 #GoogleCloud #AIAgents #AgentSkills
2116
Casey West @caseywest.bsky.social · 30/07/2026
The thirty-day retention everyone quotes for Grounding with Google Search went stale last February. It's up to 3 days of debug logs now. The old thirty days moved to Maps grounding—same provider, opposite posture: caseywest.com/governed-gro...
caseywest.com
Governed Growth, Part 2: The Retention Window That Quietly Shrank
On February 23, 2026, Google quietly made the retention terms for Grounding with Google Search better: the window dropped from thirty days to up to three, and what's kept narrowed from your prompts an...
000
Casey West @caseywest.bsky.social · 29/07/2026
Your agent can't really search the web. It guesses from training, or scrapes a snippet. gemini-search-mcp gives it one tool: a real Google search, run through Gemini, answered with the sources it used. One install, one static binary: github.com/cwest/gemini-search-mcp
000
Casey West @caseywest.bsky.social · 28/07/2026
Picked the small Gemini model for my search tool. Two newer models shipped, so I re-ran the eval—four models, one run. The small one still wins. The biggest came last on everything, at ~140x the cost. caseywest.com/building-gem...
caseywest.com
Building gemini-search-mcp: Grounding, Citations, and Choosing a Gemini Model
The builder's companion to Governed Growth—the same reference repo from the implementer's chair. Why I built a one-tool MCP server to hand an assistant a reliable, governed Google search, how Gemini g...
310
Casey West @caseywest.bsky.social · 22/07/2026
"We approved the model" and "we approved the model's web search" are two different decisions. Most platforms treat them as one—so one line in a PR opens a data path nobody chose. Part 1 of a 5-piece run. Follow along: caseywest.com/governed-gro...
caseywest.com
Governed Growth, Part 1: Better Together, Governed Apart
Opening a platform to more teams and more models—first-party and partner, such as Gemini and Claude—quietly assumes those capabilities share one trust boundary. They don't. Multi-model expansion is a ...
000
Reposted by Casey West
Remigiusz Samborski @rsamborski.bsky.social · 13/07/2026
Casey analyzes and deconstructs the viral case of a coffee shop that was run by AI agents as an experiment. How can we avoid similar mistakes?
121
Casey West @caseywest.bsky.social · 12/07/2026
They swapped the model mid-experiment. The café failed the same way. So the model wasn't the problem. It shipped with no spend cap and no approval gate—the seatbelts you install on day one. The proof and the five fixes: www.linkedin.com/pulse/ai-did...
linkedin.com
The AI Didn't Fail. The Deployment Did.
The Infographics Show, one of the biggest explainer channels on YouTube with an audience in the millions, just put out a video titled "Google Gemini Ran a Real Business for a Month and It Almost Bankr...
011
Casey West @caseywest.bsky.social · 01/07/2026
My `gh` calls started failing: rate limited. Weird. I'd barely made any. Turns out GitHub meters REST and GraphQL on two separate budgets, and agents drain GraphQL while REST sits full. Route across both, double your headroom. caseywest.com/github-gives...
caseywest.com
GitHub Gives You Two Rate Limits. You're Only Using One.
GitHub's REST and GraphQL APIs have separate hourly rate limits—two budgets, not one. Most automation hammers a single bucket while the other sits full. Here's how to see both, why agents drain GraphQ...
011
Casey West @caseywest.bsky.social · 09/06/2026
Claude Fable 5 shipped today. By lunch it was running in my own GCP project, on data terms I'd actually read. One undocumented 403 stands in the way. Walkthrough + trade-offs, stated plainly: www.linkedin.com/pulse/claude...
linkedin.com
Claude Fable 5 dropped today. Here's how to try it without gambling your company's data.
Anthropic released Claude Fable 5 this morning. By this afternoon it was the default model in my tooling, running inside my own Google Cloud project, under data terms I'd actually read.
100
Casey West @caseywest.bsky.social · 02/11/2025
Presenting The Agentic Manifesto: a new model for the era of autonomy. The old SDLC can't govern emergent AI. We need a new playbook. We call it the 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗗𝗲𝗹𝗶𝘃𝗲𝗿𝘆 𝗟𝗶𝗳𝗲𝗰𝘆𝗰𝗹𝗲 (ADLC). Read the manifesto: agenticmanifesto.ai
020
Casey West @caseywest.bsky.social · 27/10/2025
I analyzed a year of peer "thank you" notes, expecting fluff. I was completely wrong. Over 90% weren't just compliments; they were solid, admissible evidence of high performance. Next time you thank a colleague, don't just be nice. Be specific. Give them evidence they can use for their career.
000
Casey West @caseywest.bsky.social · 20/10/2025
My `archon-gemini-cli-extension` is out! 🚀 It turns 𝗚𝗲𝗺𝗶𝗻𝗶 𝗖𝗟𝗜 into 𝗮 𝘀𝘁𝗮𝘁𝗲𝗳𝘂𝗹 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗺𝗮𝗻𝗮𝗴𝗲𝗿 𝘃𝗶𝗮 𝘆𝗼𝘂𝗿 𝗔𝗿𝗰𝗵𝗼𝗻 𝘀𝗲𝗿𝘃𝗲𝗿. Use it to: → 𝗖𝗼-𝗮𝘂𝘁𝗵𝗼𝗿 𝗣𝗥𝗗𝘀 → 𝗘𝘅𝗲𝗰𝘂𝘁𝗲 𝗱𝗲𝘃 𝗽𝗹𝗮𝗻𝘀 → 𝗠𝗮𝗻𝗮𝗴𝗲 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 #AIdev github.com/cwest/archon...
github.com
GitHub - cwest/archon-gemini-cli-extension: An Archon extension for the Gemini CLI
An Archon extension for the Gemini CLI. Contribute to cwest/archon-gemini-cli-extension development by creating an account on GitHub.
000
Casey West @caseywest.bsky.social · 20/10/2025
@kweinmeister.bsky.social's 𝗽𝗿𝗶𝗺𝗲𝗿 𝗼𝗻 𝗽𝗼𝗹𝘆𝗴𝗹𝗼𝘁 𝗮𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲𝘀 𝗶𝘀 𝗿𝗲𝗾𝘂𝗶𝗿𝗲𝗱 𝗿𝗲𝗮𝗱𝗶𝗻𝗴. 𝗛𝗲 𝘁𝗲𝗲𝘀 𝘂𝗽 𝗺𝘆 𝗽𝗼𝘀𝘁 𝗳𝗼𝗿 𝗮 𝗥𝘂𝘀𝘁 𝗶𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 𝘁𝗼 𝘀𝗼𝗹𝘃𝗲 𝗣𝘆𝘁𝗵𝗼𝗻'𝘀 𝗰𝗿𝗶𝘁𝗶𝗰𝗮𝗹 𝗖𝗣𝗨 𝗯𝗼𝘁𝘁𝗹𝗲𝗻𝗲𝗰𝗸𝘀. ⚡ #𝗥𝘂𝘀𝘁 medium.com/google-cloud...
medium.com
Site Maintenance
010
Casey West @caseywest.bsky.social · 19/10/2025
On the road, folks ask where to learn AI. Here's my answer: Google Skills, your all-in-one platform for AI learning and so much more. New? Get 150 free credits!💪 #AI cloud.google.com/resources/of...
cloud.google.com
Google Skills: Skill Up in AI
Learn more about Google Skills, receive learning credits, and more.
000
Casey West @caseywest.bsky.social · 18/10/2025
I wrote this a decade ago to help managers build strong, independent teams. It's called 𝗧𝗵𝗲 𝗛𝗲𝗹𝗽 𝗧𝗶𝗺𝗲𝗿: a technique I’ve used with a lot of teams to help them balance independence with collaboration. A tool for mentorship, not just efficiency. 🧠⚙️ #EngLead caseywest.com/the-help-timer
caseywest.com
The Help Timer
Wait an hour to ask for help. No more, no less.
000
Casey West @caseywest.bsky.social · 12/10/2025
𝑁𝑒𝑤 𝐴𝐼 𝑎𝑔𝑒𝑛𝑡 𝑓𝑟𝑎𝑚𝑒𝑤𝑜𝑟𝑘 𝗔𝗿𝗰𝗵𝗼𝗻 is 𝘳𝘢𝘥! 💥 Just had its first release. I've been trying it with 𝗚𝗲𝗺𝗶𝗻𝗶 𝗖𝗟𝗜 for 𝗚𝗲𝗺𝗶𝗻𝗶 𝟮.𝟱 models, local dev, & smart RAG. Impressive and ready for you to explore! #AI #AgenticAI #Gemini github.com/coleam00/Arc...
github.com
GitHub - coleam00/Archon: Beta release of Archon OS - the knowledge and task management backbone for AI coding assistants.
Beta release of Archon OS - the knowledge and task management backbone for AI coding assistants. - coleam00/Archon
000
Casey West @caseywest.bsky.social · 10/10/2025
Come to my half day, hands on workshop "From Prompt to Action: Building Production-Ready AI Agents" in Istanbul on November 7th! Sign up to leave with the skills to build, deploy, and scale dynamic, production-grade AI solutions. rsvp.withgoogle.com/events/devel...
rsvp.withgoogle.com
Developer Day Istanbul
Developer Day Istanbul
000
Reposted by Casey West
Remigiusz Samborski @rsamborski.bsky.social · 06/10/2025
🧹 This Halloween Accelerate AI with Cloud Run lands in Munich! 🇩🇪 Join @kweinmeister.bsky.social, Shir Meir Lador, @caseywest.bsky.social and me to learn that Cloud Run,Gemini,GPUs aren't so scary after all!😱 👉Register: goo.gle/accelerate-a... Seats are limited! #AI #CloudRun #GoogleCloud #Munich
031
Casey West @caseywest.bsky.social · 04/10/2025
You have to use the right tool for the job. But what if your AI agent's tools can't be written in the right language? I just published a 2,800-word guide on building decoupled, polyglot AI agents that can actually scale. caseywest.com/building-sca... #AIArchitecture #Rust #Go #Python
caseywest.com
Building Scalable AI Agents: A Deep Dive into Decoupled Tools with ADK, MCP, and Cloud Run
A deep-dive playbook for building scalable, enterprise-grade AI agents. Learn how to move beyond monolithic designs by decoupling tools with a powerful microservices architecture using ADK, the Model ...
010
Reposted by Casey West
Remigiusz Samborski @rsamborski.bsky.social · 30/09/2025
Dublin 🇮🇪 we're coming! Learn to build AI agents and deploy your MCP servers to production scale. Register: goo.gle/accelerate-ai-dublin Seats are limited! Shir Meir Lador @kweinmeister.bsky.social @caseywest.bsky.social #AI #AIagents #MCPServers #CloudRun #Workshop @GoogleCloudTech #DublinEvents
012
Casey West @caseywest.bsky.social · 24/09/2025
Big news from the 2025 DORA Report on AI in dev! 🚀 My article breaks down 8 crucial shifts: AI as an amplifier, platform engineering is key, and it's even boosting authentic pride. It's a systems game. www.linkedin.com/pulse/ai-sof...
linkedin.com
AI in Software Development is Not What You Think: 8 Revelations from the 2025 DORA Report
Remember last year's DORA report? The one that dropped a bombshell finding: AI was actually slowing down software delivery and making things less stable. It was a surprising, counterintuitive result t...
000
Casey West @caseywest.bsky.social · 28/06/2025
I’m about eight years late to Strava, but if you want to cheer each other on using that social network, let’s connect! www.strava.com/athletes/cas... 🥾👟🚲🧗🧘‍♂️🏋️
strava.com
Casey West | Strava Runner Profile
Casey West is a runner from United States. Join Strava to track your activities, analyze your performance, and follow friends. Strava members can plan routes, participate in motivating challenges, and...
000
Casey West @caseywest.bsky.social · 11/06/2025
👋🏼 See you in two weeks. 🥾 I'm going to hike the Alta Via 1 in the Dolomites. ⛰️
010
Casey West @caseywest.bsky.social · 10/06/2025
Impactful use of AI doesn’t help me do 𝘢 𝘭𝘰𝘵 𝘮𝘰𝘳𝘦, it helps me do 𝗮 𝗹𝗼𝘁 𝗯𝗲𝘁𝘁𝗲𝗿, much more often. www.linkedin.com/pulse/stop-chasing…
000
Casey West @caseywest.bsky.social · 08/06/2025
Is your team's AI just creating "10x mediocrity?" We're all chasing productivity, but it's a dangerous trap if you're solving the wrong problems. I wrote about how to shift from raw output to real impact. www.linkedin.com/pulse/stop-c...
linkedin.com
Stop Chasing AI Productivity. Start Driving Real Impact.
We’ve all seen it. An eager engineer, proud of their newfound AI prowess, triumphantly presents a pull request with thousands of lines of perfectly generated code.
010
Casey West @caseywest.bsky.social · 02/06/2025
Want AI to make you think more, not less? My latest LinkedIn article breaks down how systems thinking, metacognition, & evolving your prompts can unlock true AI collaboration and "system masterpieces." Full piece here: www.linkedin.com/pulse/beyond... #AI #FutureOfWork #Innovation
linkedin.com
Beyond Basic Commands: Mastering AI with Systems Thinking, Metacognition, and Meta-Prompting
The AI revolution isn't just about the technology; it's about transforming how we think, interact, and create. As Large Language Models (LLMs) and other AI tools become increasingly integrated into ou...
011
Casey West @caseywest.bsky.social · 30/05/2025
If you owned the domain name yolo.engineering what would you forward it to right now?
000
Casey West @caseywest.bsky.social · 29/05/2025
An absolute rule is wrong 100% of the time.
100
Casey West @caseywest.bsky.social · 28/05/2025
Tired of LLM prompts that are just "meh"? 🫠 My secret to getting them to "elite agent" level is out! I'm basically giving away how I get AI to help me build better AI. Check out how I built a "Meta Prompt Collaborator" to ditch tedious prompt writing: caseywest.com/unlock-elite...
caseywest.com
Unlock Elite Agents: The Art of Evolving LLM Prompts into System Masterpieces
Explore the evolution from basic LLM interactions to crafting sophisticated System Instructions, and discover how meta-prompting—using AI to refine AI prompts—can unlock more powerful and collaborativ...
001
Casey West @caseywest.bsky.social · 16/05/2025
Learnings aren't a thing. You hear it in meetings, but "learning" is usually an uncountable noun, like "information" or "knowledge." We don't say "knowledges." "Learnings" is jargon. Use clearer language like "lessons learned," "insights," or just "lessons." Your communication will be stronger.
000
Casey West @caseywest.bsky.social · 13/05/2025
What are you using for an RSS reader in 2025? I'm mostly apple ecosystem.
110
Casey West @caseywest.bsky.social · 07/05/2025
Untangling LLM evaluation! 🧐 My new LinkedIn article tackles common mistakes when assessing LLMs for specific apps. Hint: it's not just about those big leaderboard scores. Check it out: www.linkedin.com/pulse/demyst... #AI #LLM #Evaluation
Clean illustration of a magnifying glass hovering over a stylized neural network. Some pathways are bright, others are tangled leading to question marks. Professional, tech-focused.
000
Casey West @caseywest.bsky.social · 05/05/2025
Gemini hurt my feelings (on my personal account) 😂
Gemini cuts me off with a rate limit error message.
110
Casey West @caseywest.bsky.social · 02/05/2025
Has anyone used a Naya Create keyboard? I’m no keyboard expert, and I’m not looking for clacky keys. I’m curious about their portability and personalization. It looks something like this. naya.tech
Naya Create Keyboard from the OG Kickstarter
000
Casey West @caseywest.bsky.social · 28/04/2025
Tired of juggling pyenv, venv, pip-tools, and pipx? 🤹‍♂️ uv aims to be the one tool to rule them all for Python dev environments. My latest post explores building a fast, simple workflow around it. Includes migration tips! ✨ #Python #uv #Workflow caseywest.com/taming-the-p...
caseywest.com
Taming the Python Hydra: A Modern Dev Environment with uv
Stop wrestling with Python versions and dependencies. Learn how to build a fast, simple, and unified development environment using uv
010
Casey West @caseywest.bsky.social · 24/04/2025
8 years at Google Cloud today! 🎉 Feeling incredibly grateful for the journey. I shared some personal reflections and thoughts on what I'm up to now in this article: www.linkedin.com/pulse/my-goo... #Googleversary #GoogleCloud #DeveloperAdvocacy
020
Casey West @caseywest.bsky.social · 23/04/2025
Revamped my decade-old Jekyll site -> Astro in just 3 days! 🤯 Partnered w/ Google Gemini for planning, coding, and even building custom TS tools. The future is fast! 🚀 Case study on this AI partnership: caseywest.com/ai-driven-development…
020
Casey West @caseywest.bsky.social · 20/04/2025
Today I did my first real “Vibe Coding” session to update my old and busted personal site from Jekyll → Astro. I’m using VS Code, Roo Code, and Gemini 2.5 Pro Preview. So worth the $1. Give me a bit of time to migrate the writing I want to keep. Then I’ll write about how I’m doing this, too. 😁
A message box with a dark background displays text: "APRIL 19, 8:14 PM Replace the Twitter icon and link with Bluesky. My handle is @caseywest.bsky.social Tokens: ↑ 320.7k ↓ 17.1k API Cost: $1.0582"
010
Casey West @caseywest.bsky.social · 01/01/2025
Happy New Year! 🎉 My 2025 resolution is to be 3% better to keep my value against inflation.
010