אודות פרק זה
“Most assays don’t fail because they’re wrong… they fail because they’re fragile.”
This time, Markus Gershater and Phil Kay sat down with Rob Howes (Head of Biology & DMPK at Charles River) to unpack what REALLY happens when you scale biology - and why most teams underestimate how quickly things fall apart.
Rob’s been working with multi-dimensional experimentation since the early 2000s (back when most of us were still painfully tweaking one variable at a time). And the contrast is pretty stark:
Some standout lines:
🔹 “We used to optimise for the highest signal. Now we optimise for something that won’t break.”
🔹 “If your data isn’t high quality, your AI model won’t be either. Garbage in, just faster garbage out.”
🔹 “Why run every experiment, if you can predict most of them?”
