Lab work is "high mix, low volume," like driving, making it hard to automate. Traditional automation is like a subway: efficient but inflexible. AI enables "autonomous" labs, akin to Waymo cars, that handle the vast variability of experiments, which constitutes 99% of lab work.

The lab of the future should abandon human-centric design and instead emulate a data center. This means prioritizing density, energy efficiency, and automation to maximize the generation of scientific data ('tokens') around the clock, with minimal human intervention.

Instead of pursuing 100% automation, Lila's system treats complex manual tasks as API calls routed to human technicians. This pragmatic approach keeps humans 'below the API line,' seamlessly integrating necessary manual work into a software-driven workflow without halting automation.

Scientific research is being transformed from a physical to a digital process. Like musicians using GarageBand, scientists will soon use cloud platforms to command remote robotic labs to run experiments. This decouples the scientist from the physical bench, turning a capital expense into a recurring operational expense.

Traditional "flexible" lab design pre-engineers for every possible future scenario, which is expensive and rigid. A smarter approach is "adaptability": consciously designing pathways and leaving space for future technology without over-investing in systems that may quickly become obsolete.

Unlike pre-programmed industrial robots, "Physical AI" systems sense their environment, make intelligent choices, and receive live feedback. This paradigm shift, similar to Waymo's self-driving cars versus simple cruise control, allows for autonomous and adaptive scientific experimentation rather than just repetitive tasks.

When automating lab processes, the primary challenge is not adapting to new scientific methods but scaling the infrastructure to handle the massive, 24/7 flow of data from instruments and process logs. This requires a robust data management strategy from the outset.

To create a complex automated science platform, first build modular tools that human experts use in a manual workflow. Observe their process to identify bottlenecks and needed components (e.g., a stability test). Then, incrementally build agents to automate the orchestration of these proven tools.

The primary value of AI in bioprocessing is not just automating tasks, but analyzing process data to predict outcomes. This requires a fundamental shift in capital equipment design, focusing on integrating more sensors and methods to collect far more granular data than is standard today.