Lab
AI tools, creative workflows, developer deep-dives, and behind-the-scenes from RAXXO Studios.
62 articles
5 Claude API Errors That Cost Me Money (And How I Trapped Them)
Retry storms turned 1 timeout into 340 duplicate calls billed in 90 seconds Infinite tool loop ran 1,200 iterations before I noticed at 2am Partial...
The 1M Context Window vs Prompt Caching: When to Use Which
1M context costs full price on every query, caching cuts repeated tokens to 1/10 Use 1M for one-shot deep dives, caching for repeated calls against...
Shopify Admin API From Claude: 5 Patterns I Use Across 15 Repos
Product create via Admin API beats clicking through the dashboard every time Blog post POST with metafields saves 12 minutes per article Metaobject sync keeps...
What I Run as Cron Jobs in My Claude Stack: 6 Automations
6 cron jobs run my studio publishing pipeline on a 2 EUR/month server Auto-syndicate posts to 4 channels at 09:00 daily Blog-index rebuild runs every...
Claude Files API in Production: 5 Patterns for Document Workflows
Files API replaced my 40KB inline blobs with reusable file IDs across requests Citation grounding cut hallucinated quotes to near zero in 200 test runs...
How I Auto-Publish a Blog Article Per Day With GitHub Actions and Claude
Daily blog automation runs free on GitHub Actions cron at 06:00 UTC Topic queue file plus dedup check against the live blog index Single Anthropic...
Building Multi-Agent Workflows With Claude: A Solo Studio Playbook
Three parallel writer agents drafting at once cut my blog turnaround from 6 hours to 90 minutes Research agents run in parallel then merge through...
5 Anthropic Prompt Caching Patterns That Cut My API Bill 70%
System-prompt caching alone cut repeat-call costs by half Tool definitions cache separately, perfect for agent loops Conversation history caching pays off after turn three 1-hour...
Opus 4.8 vs Sonnet vs Haiku: How I Route Work in 2026
Which Claude model for which job in 2026, with Opus 4.8's flat price, cheaper Fast mode, and abstention as the deciding factors. My routing table.
Opus 4.8 in Practice: A Real Coding Session
A field report from one real agentic coding session on Opus 4.8: where it asked instead of guessing, caught its own bug, and held the...
Claude Opus 4.8 Is Here: Everything That Changed
Claude Opus 4.8 landed May 28 with a 4x drop in unremarked code bugs, a 3x cheaper Fast mode, Dynamic Workflows, and the same price...
5 Anthropic SDK Streaming Patterns I Use in Production
Token paint with backpressure stops fast streams from crashing slow UIs Partial tool_use blocks let you show "querying X..." 200ms into a call AbortController cancels...