Why Life-Science Labs Still Run on Spreadsheets — and What That Costs
The average biotech research team passes experiment data through at least four disconnected tools before it is ever recorded in a durable place. Here is why that keeps happening.
Research Notes
Research Notes
Thinking from the Cypher AI team on lab informatics, experimental reproducibility, and what it actually takes to build software for scientists.
Designing the Lab OS: How AI Agents Write Tools Scientists Actually Need
Why "chat with your data" misses the actual problem in a research lab — and what a useful unit of AI agency in a scientific workflow actually looks like.
Protocol as Code: Making Biology Reproducible by Default
What software engineering learned about reproducibility in the 2010s, biological research is beginning to learn now. The tools are different; the principle is identical.
From ELN to Experiment OS: What the Next Decade of Lab Software Looks Like
Electronic lab notebooks were a step forward. But as research grows more computational, labs need more than a better notebook — they need an operating system.