What it does

Blaise is an LLM-native workspace for small-molecule design. It’s built around critique, not ideation: SAR reasoning, matched-pair style thinking, property trade-offs, and “what would I make next?” decision support.

The core output you want isn’t a molecule — it’s a defensible recommendation with assumptions, failure modes, and the smallest next experiment that de-risks the program.

Who it’s for

• Medicinal chemists running design cycles and triaging ideas under tight timelines.
• Computational chemists who want model outputs translated into decision-ready language.
• Project teams who need consistent rationale across meetings, handoffs, and months.

What we believe

• Drug discovery is mostly rejection. Speed comes from killing weak ideas early.
• “Potency” isn’t a scalar. It’s series-conditional, mechanism-conditional, and property-coupled.
• Good taste is operational. If you can’t explain it, you can’t scale it across a team.
• Constraints are the product. ADME, safety, and synthesis are not footnotes — they’re the design space.

This blog

We publish the underlying reasoning: concrete heuristics, failure analyses, series strategy notes, and the occasional product deep dive. The theme is always the same — turning messy evidence into sharp decisions.

If you like writing that sounds like a lab meeting (but tighter), you’ll feel at home here.