Work in progress
Experiments are prototypes and half-built ideas. Some ship as products. Others fail and teach us what to skip next time.
Context-Aware Routing
Can a lightweight classifier determine which AI agent should handle a given task?
Testing routing accuracy and latency on a classification model. Results will be published with test methodology.
Streaming Markdown Renderer
A renderer that displays AI-generated markdown as it streams, without layout shift.
The challenge is parsing partial markdown correctly while keeping the DOM stable. Covering edge cases in active test suites.
Design Token Compiler
A build tool that takes a single JSON token definition and outputs CSS, Tailwind, and Figma variables.
The goal is one source of truth for all brand tokens across platforms.
Edge Function Auth Gateway
Moving preview authentication from API routes to edge functions for lower latency.
Targeting measurable response time reduction. Latency numbers will be published once testing completes.
Specific metrics will be published with methodology documentation.
How we experiment
Question
Start with a specific, answerable question. No vague exploration.
Prototype
Build the smallest thing that could answer the question. Days, not weeks.
Validate
Run it against real data. Measure. Compare to the baseline.
Decide
Ship it, iterate, or kill it. No zombie experiments.