Alex Patterson

Want to get more @codingcat.dev in your life, subscribe to our newsletter! buttondown.com/codingcatdev...

Sending sensitive patient data to a cloud LLM API just to parse a voice request is a massive HIPAA compliance headache. With local Speech-to-Text and a 12B Gemma model, you can execute tool calls 100% offline on-device. Breakdown in the reply 👇

Instead of letting an agent guess which vector chunk matters, start with two things in the config: a sharp dataset purpose and a focused GROQ projection. That produces a clean outline for direct entry-path reads via Context MCP. See it in the reply 👇

Podcast S1E18 with Lauren Lee (@laurenlee.bsky.social): Fostering Developer Education We sit down with Lauren Lee (LoLoCoding) to talk about fostering developer education. We review how Lauren switched from being a teacher to become the Director of Community at Puppet

CodingCatDev Deep Dive: Cloudinary in Jamstack using Webp Create a single source for all of your posting needs, using the best format for the browser!

Offline On-Device AI Agents with Gemma & Local Tool Calls (feat. @shivaylamba.bsky.social) How do you run an agentic healthcare workflow without sending private patient data to the cloud? Run local Speech-to-Text + a 12B Gemma model on-device to execute pharmacy tool calls 100% offline.

Podcast S1E17 with Dan Ribbens: Payload CMS We chat with Dan Ribbens about Payload CMS a self-hosted, headless JavaScript CMS.

Podcast S1E16 with Sam Bhagwat (@calcsam.bsky.social): Delivering a Content Mesh On this podcast we talk with Sam Bhagwat, Cofounder & Chief Strategy Officer at Gatsby. We dive into Gatsby 3.0 and why Jamstack is allowing us

Running AI agents in production comes down to unit economics and governance. Combining a lightweight `initial_context` outline with scheduled build diffs keeps token costs low and catches source drift early. Production notes in the reply 👇

Podcast S1E15 with Shawn Wang (swyx): What's up with Svelte On this podcast we talk with Shawn Wang also known as swyx. We dive into what is happening with Svelte, the Svelte Society, and Svelte Radio.

CodingCatDev Deep Dive: Use Firestore to Build Hugo Content Use a frontend app (Angular), to update a Firestore Backend, trigger Firebase Function, while maintaining git commits, and buld/deploy your Hugo site.

Podcast S1E14 with Patrick Martin (@pcnmartin.bsky.social): Firestore for Unity and C++ We talk with Patrick Martin from Firebase all about game development and how Firebase's Firestore is now available in Beta.

Podcast S1E13 with Scott Tolinski (@tolin.ski): Creating a Learning Platform We talk with Scott Tolinkski of LevelUpTutorials all about his past and why he started his Learning Platform.

Anthropic's Form S-1 ($ANT) exposes $518B in take-or-pay cloud commitments to GCP, AWS & Azure. How 7 CEOs built a circular AI loop (and why 37% of your S&P 500 401(k) is in 1 trade): https://codingcat.dev/post/anthropic-s1-ipo-circular-ai-economy-etf-risk

What happens when someone edits a pricing or policy doc with an unverified claim? In standard RAG, the vector index ingests it immediately. With Sanity Knowledge Bases, the conflict is flagged before it touches agent state. Demo in the reply 👇

Shipping an MCP knowledge pipeline to production requires three guarantees: low token overhead, source drift protection, and strict read-only auth. Never let unvetted doc edits silently mutate agent answers. Production patterns in the reply 👇

How do you keep AI agent context windows lean at runtime? Shape the dataset at build time. Use a narrow GROQ projection to strip UI metadata and compile a compact dataset outline so agents retrieve exact entry paths on demand. Details in the reply 👇

Podcast S1E12 with Joel Varty (@joelvarty.com): Agility CMS Joel Varty talks all about Agility CMS. Where DXP fits into the Jamstack.

CodingCatDev Deep Dive: NPM Tips and Tricks Adding NPM Globally, setting default values for packages, and stopping the dreaded sudo calls.

Podcast S1E11 with Jason Lengstorf (@jason.energy): Moving from WordPress to Jamstack Talking with Jason Lengstorf about how to use Wordpress to power the backend to any of your Jamstack sites.

Stop Wasting Tokens: Give AI Agents a Dataset Outline. Keep AI agents accurate and token-efficient by replacing broad similarity searches with a lightweight dataset outline and direct entry-path retrieval. Start with a clear purpose and a narrow Groq proj

Podcast S1E10 with Aaron Frost, Ali Spittel (@alispit.tel), Kelly Vaughn (@kvlly.com), Tyler McGinnis (@tyler.ui.dev): Creating a Startup Creating a Startup is hard! Learn from the experts who have created great startup companies. Everything from an LLC to Stripe's new C-Corp tool...

Podcast S1E9 with Elizabeth Wilson (@wilsones13.bsky.social), Brett Fitzgerald: SAFe is SCRUM on Steriods Chatting with Elizabeth Wilson and Brett Fitzgerald about Agile, SCRUM and SAFe practices when building products.

CodingCatDev Deep Dive: VSCode Extension Peacock Subtly change the color of your Visual Studio Code workspace. Ideal when you have multiple VS Code instances, use VS Live Share, or use VS Code's Remote features, and you want to quickly identify your editor.

Podcast S1E8 with Christy Ennis-Kloote, Zoë Wilson (@zoelwilson.bsky.social), Elizabeth Bush: Design for Developers Are you struggling to become a designer for web development? You don''t need formal web development training to be successful. Learn from the design group at OST how...

Broad vector similarity searches burn tokens on noisy chunks and miss structured context. Instead of dumping raw RAG hits into the prompt, give AI agents a lightweight dataset outline with direct entry-path retrieval. Full breakdown in the reply 👇

Podcast S1E7 with Ovi Negrean, Vlad Hosu (@vlad.socialbee.com): SocialBee with Ove and Vlad A discussion with SocialBee founders Ovi Negrean and Vlad Hosu. We talk about the SAAS product SocialBee and how it is used for people to save time posting to socials. The architecture and...

Documentation never stays static. Instead of rebuilding your entire knowledge index blindly, Sanity Knowledge Base runs scheduled weekly or monthly refreshes, diffs sources against your last build, and flags material changes. See it in the reply 👇

Why pay token latency at runtime to reconcile conflicting RAG chunks? Moving knowledge compilation to build time with deterministic GROQ projections and Context MCP guarantees your agent reads verified state. Architecture in the reply 👇

Podcast S1E6 with Colby Fayock (@colbyfayock.com): Jamstack Handbook by Colby Fayock Talking with Colby Fayock we dive into Jamstack security, Jamstack ecommerce, and much more!

CodingCatDev Deep Dive: Ionic Icons How can you use Ionic Icons, anywhere and everywhere?

Podcast S1E5 with Chris Biscardi (@chrisbiscardi.bsky.social): MDX with Chris Biscard MDX is an authorable format that lets you seamlessly write JSX in your Markdown documents. We talk about MDX and the technology surrounding MDX with Chris Biscardi.

Podcast S1E4 with Guillermo Rauch (@rauchg.blue): Next.js 10 Talking with Guillermo Rauch about Next.js 10 and Vercel. Including plans for 2021.

What happens when you feed an AI Knowledge Base a Sanity dataset, crawled GitHub pages, and a custom lore file with overlapping facts? The ahead-of-time build flags the conflict side by side before your agent ever sees it. Live demo in the reply 👇

Podcast S1E4 with Tim Benniks (@timbenniks.dev): Building your Own SaaS Tim teaches Alex how to make your own SaaS using Supabase, Algolia and Nuxt.

CodingCatDev Deep Dive: Cloud Function GitHub Update Changing git libraries

Podcast S1E3 with René Brandel (@renebrandel.bsky.social): AWS Amplify Admin UI with René Brandel Talking with René Brandel about AWS Amplify Admin UI. How you can get started with AWS Amplify without ever writing a line of code.

Traditional vector RAG dumps contradictory chunks into your prompt and expects the LLM to guess which one is current. Compile verified knowledge at build time instead of debugging hallucinations at query time. Breakdown in the reply 👇

Podcast S1E3 with Simon Grimm (@galaxies.dev): Expo from a Web Developers View Dive into the seamless world of cross-platform app development as we unravel the magic of React Native with Expo, combining power and simplicity for an unparalleled mobile experience.

How do you give an agent access to PDFs, spreadsheets, and Portable Text without paying the token tax on every turn? Hold a lightweight topic outline in `initial_context` and fetch specific entries on demand. Architecture in the reply 👇

Podcast S1E2 with Brian Li: Kinsta Talking with Brian Li all about WordPress hosting provider Kinsta and why they chose to utilize Google Cloud Platform.

CodingCat.dev Deep Dive: AJ is Loving Sticker Mule I have tried several sticker places and the images always come back terrible and the customer service stinks. Then I found Sticker Mule and everything has been like a dream!

Podcast S1E2 with Thomas G. Lopes (@thomasglopes.com): Open Source Svelte Magic: Melting Minds through Melt UI! Thomas teaches Brittney and Alex how to use Melt and how Runes can be used to improve builders.

Sanity Knowledge Base: Lower Token Costs, Higher Accuracy. Unify text, PDFs, spreadsheets, and crawled documentation in Sanity’s topic tree. Fetch only the entries each question needs to reduce token costs and improve accuracy, with scheduled refreshes th

Podcast S1E1 with Lee Briggs (@brig.gs): Infrastructure as Code with Pulumi Talking with Lee Briggs about Infrastructure as Code (IaC). Lee explains what IaC means and how Pulumi is working on a solution that allows you to stay in the same language as your code.

Similarity search is a leaky abstraction for structured knowledge. Instead of hoping top-k vector chunks have the right context, give your agent a purpose-steered outline via `initial_context` and read exact Markdown paths. Architecture in the reply 👇

Podcast S1E1 with Jordan Powell (@jordanpowell.bsky.social): Angular 17, does SSR/SSG really work? Join Jordan and Alex as they try to get Angular 17 SSG working and talk about SSR.

Stuffing entire PDFs, spreadsheets, and docs into an LLM prompt burns token budgets and degrades accuracy. Sanity Knowledge Base unifies disparate sources into a single topic tree so agents only load what a query needs. Breakdown in the reply 👇

Naive vector RAG chunks raw documents and forces your LLM to resolve contradictions at query time. Hardcoding Markdown in system prompts breaks when content updates. Ahead-of-time Knowledge Base compilation fixes both. Full walkthrough in the reply 👇