Evaluate Nemotron vision before moving document extraction
A practical evaluation plan for Nemotron 3.5 Super VL on DigitalOcean: screenshot accuracy, token costs, caching, privacy, and deployment limits.
Explore the journal
AI Development
Provider setup, testing boundaries, and production checks for useful AI workflows—so features stay grounded in verified behavior, not guesswork.
Build AI features with clearer boundaries.
Showing 6 of 9 articles
Run OpenCode from Your iPad with a Private Cloud Workspace
Use OpenCode in Safari while your project lives on a DigitalOcean server. Set up private HTTPS access, keep work between sessions, and understand the costs.
Choose models per request with a measured routing test
Test DigitalOcean's redesigned Inference Router against a fixed model, check the full request cost, and keep a safe rollback before changing production.
Check prompt cache savings before switching inference providers
A practical preflight for measuring cached input, protecting tenant boundaries, and comparing the full cost of serverless inference.
Route support tickets with Jev on DigitalOcean
Call Jev through DigitalOcean Serverless Inference, estimate token costs, and test a review workflow before enabling automatic support-ticket routing.
Call GPT-6 Astra through DigitalOcean with the Responses API
Call GPT-6 Astra through DigitalOcean with the Responses API. Check access, create a key, send a request and compare costs before adding routing or agents.