Articles by Grizzly Peak Software

Gumroad + Agents: Automating Product Launches End-to-End
AI
Gumroad + Agents: Automating Product Launches End-to-End

Last month I launched a new digital product on Gumroad — a 40-page PDF guide on building production-ready API gateways. The entire launch process, from...

Read Article
Building Resilient Agents That Don't Break When APIs Change Overnight
AI
Building Resilient Agents That Don't Break When APIs Change Overnight

It was a Tuesday morning. I was drinking coffee and watching moose tracks fill with fresh snow outside my cabin window when my phone started buzzing. The...

Read Article
Integrating Google CLI + LLMs: Automating My Alaskan Homesteading Logs
AI
Integrating Google CLI + LLMs: Automating My Alaskan Homesteading Logs

Most people hear "homesteading in Alaska" and picture some guy with an axe and a flannel shirt who has never touched a computer. And honestly, some days...

Read Article
Craftsman Meets Code: Building Furniture with Parametric Design and LLMs
AI
Craftsman Meets Code: Building Furniture with Parametric Design and LLMs

I broke three pieces of walnut last winter trying to build a bookshelf.

Read Article
Gumroad Factory 2.0: Automating Cover Art, Landing Pages, and Emails
AI
Gumroad Factory 2.0: Automating Cover Art, Landing Pages, and Emails

I sold my first digital product on Gumroad in 2023. It was a PDF about API gateway patterns, the cover looked like it was designed in PowerPoint by...

Read Article
Fixing Claude's Slow Response Nights: Local Fallbacks & Hybrid Setups
AI
Fixing Claude's Slow Response Nights: Local Fallbacks & Hybrid Setups

If you've spent any serious time building with Claude's API, you've noticed the pattern. Mornings on the West Coast? Snappy. Sub-two-second responses on...

Read Article
When LLMs Hallucinate Architecture: Real Enterprise Integration Nightmares
AI
When LLMs Hallucinate Architecture: Real Enterprise Integration Nightmares

Last month, a developer I mentor showed me an architecture diagram that Claude had generated for his team's new microservices migration. It was beautiful....

Read Article
AI Code Review Agents: Saving Hours on Pull Requests (My Workflow)
AI
AI Code Review Agents: Saving Hours on Pull Requests (My Workflow)

I used to spend two hours every morning reviewing pull requests. Coffee in hand, cabin quiet except for the woodstove popping, just grinding through diffs...

Read Article
Why Flash Models Are Crushing Mini Versions for Indie Hackers Right Now
AI
Why Flash Models Are Crushing Mini Versions for Indie Hackers Right Now

Last month I ran a batch classification job through three different models: GPT-4o-mini, Gemini 2.0 Flash, and Claude 3.5 Haiku. Same prompts, same data...

Read Article
AI for Craftsman Projects: Generating CNC Plans from Voice Descriptions
AI
AI for Craftsman Projects: Generating CNC Plans from Voice Descriptions

I was standing in my shop last November, staring at a piece of walnut I'd been saving for two years, trying to figure out how to turn a vague idea in my...

Read Article
Why Gemini Flash 2.0 Just Became My Daily Driver (Benchmarks + Real Builds)
AI
Why Gemini Flash 2.0 Just Became My Daily Driver (Benchmarks + Real Builds)

I'll admit it — I have a model loyalty problem. For the past year, I've defaulted to Claude for nearly everything. Writing articles, debugging code...

Read Article
One-Click Deploy Agents: Streamlining Vercel + Supabase Setups
AI
One-Click Deploy Agents: Streamlining Vercel + Supabase Setups

I got tired of the same deployment ritual. Create a new Vercel project. Connect the GitHub repo. Set up environment variables. Spin up a new Supabase...

Read Article
Why I am Betting Against Claude for Enterprise in 2027 (Hot Take + Evidence)
AI
Why I am Betting Against Claude for Enterprise in 2027 (Hot Take + Evidence)

I use Claude every day. Claude Code is my primary development tool. I built a job classifier with Claude Haiku that processes hundreds of job listings...

Read Article
o1 Pro Mode s Massive Context Window: Building 50k-Token Enterprise Specs Without Crashing
AI
o1 Pro Mode s Massive Context Window: Building 50k-Token Enterprise Specs Without Crashing

Last month I tried to feed an entire microservices architecture spec into GPT-4. The spec was about 38,000 tokens — not even that big by enterprise...

Read Article
Copilot CLI Hacks for Non-Technical Founders (Tested on My Latest SaaS)
AI
Copilot CLI Hacks for Non-Technical Founders (Tested on My Latest SaaS)

Last Tuesday I was sitting in my cabin in Caswell Lakes, Alaska, watching a founder friend struggle over Zoom. He had a Rails app deployed on Heroku, a...

Read Article
Claude Artifacts vs. Copilot Workspace: Which Wins for Rapid Prototyping in 2026?
AI
Claude Artifacts vs. Copilot Workspace: Which Wins for Rapid Prototyping in 2026?

Last Tuesday at 11 PM, I had an idea for a small tool — a dashboard that pulls job posting stats from my Grizzly Peak Software job board and shows hiring...

Read Article
Multimodal AI Wins: Using Vision Models to Debug UI Screenshots Automatically
AI
Multimodal AI Wins: Using Vision Models to Debug UI Screenshots Automatically

Last Tuesday at 11 PM, I was staring at a bug report from a user of my AutoDetective.ai site. The report said "the results page looks broken on mobile."...

Read Article
Prompt Engineering for Alaskan Winters: Offline-First AI Tools That Work Without Internet
AI
Prompt Engineering for Alaskan Winters: Offline-First AI Tools That Work Without Internet

Last January, a snowstorm knocked out my internet for four days. Not four hours — four days. The satellite dish on the roof of my cabin in Caswell Lakes...

Read Article