a16z Podcast | When (and How) Biology Becomes Engineering

Shift from Stochastic Science to Engineering: The conversation defines the transition in biology as moving from a high-risk, hypothesis-test-revise scientific model (stochastic) to a methodical design-build-test-refine engineering model.

• Science is characterized by discovery and revision; engineering is characterized by standardization and iteration.
• The goal is to reduce risk by treating biological components as "Legos" that can be mixed, matched, and standardized.

The "Lego" Analogy in Biology: Biological systems possess a natural hierarchy (atoms → molecules → cells → tissues → organs → organisms).

• Part Identification: The "science" phase involves discovering distinct biological parts (e.g., phenol rings for molecules, specific cells for modern drugs).
• Standardization: Success relies on finding parts that "click" together predictably, similar to mechanical or electrical engineering, rather than creating bespoke solutions for every project.
• Current Example: Companies like Asimov are developing standardized biological parts, moving the industry from research into construction.

Role of Interdisciplinary Engineering: Traditional engineering disciplines are increasingly applicable to biological problems.

• Mechanical engineering, electrical engineering, material science, and computer science are crossing into biology.
• Material Substitution: Principles of steel and bridges are being applied to bone and muscle.
• Academic Response: New "Bioengineering" departments have emerged in the last 10 years to train individuals with hybrid skill sets, accelerating the flow of engineering talent into biology.

Evolution of Genetic Engineering:

• Historical State: Previously viewed as "playing Boggle" (randomly mixing letters/genes to see what works).
• Current State: Technologies like CRISPR and synthetic biology platforms have transformed DNA into a true "design medium," making genetic engineering a genuine engineering discipline.

Machine Learning as an Engineering Catalyst: ML shifts biology from bespoke, decade-long discovery processes to repeatable, data-driven engineering.

• Process Standardization: ML allows for the engineering of processes where new data is simply rolled through existing systems.
• Value Shift: In an engineering context, the second asset is often more valuable than the first due to accumulated learning, reversing the traditional biotech model where chronological order determined value.
• Data Utilization: Pharma companies are increasingly viewed as "data generation companies," where false positives are as valuable as true positives for iterative learning.

Future Workforce and Infrastructure:

• Dry Lab Growth: Prediction that in 10 years, pharmaceutical companies will have "dry labs" (computational/ML teams) comparable in size to "wet labs" (pipette-based teams).
• Role Evolution: Medicinal chemists are transitioning into "drug designers" or "drug engineers," focusing on design rather than synthesis, which is increasingly outsourced to CROs (Contract Research Organizations).

Commercialization and Business Development:

• Proof of Concept: The key activation energy for selling biological platforms is demonstrating predictability and reproducibility, not just novelty.
• Predictability Metric: Buyers need to see that a designed outcome works a "significant percentage of the time," inverting the traditional "1 in 10,000" success rate of discovery biology.
• Pilot to Deal: Engineering-based companies can use OKRs and KPIs to set short-term delivery timelines, facilitating a "land and expand" strategy previously impossible in bespoke biotech.

The Compounding Effect of Platform Technology:

• Predictable Growth: Platforms that systematically improve (e.g., getting 30% better every year) follow an exponential growth curve similar to Moore's Law.
• Case Study: Illumina's success in sequencing demonstrated how demonstrating success in one context, then a second, makes the technology a universal truth by the third context.
• Long-term Vision: The ultimate goal is to make the "impossible" possible through compounding technological improvements, moving from random chance to engineered certainty.

The Apollo Mission Paradigm:

• Stepwise Engineering: Large goals (e.g., curing cancer, extending longevity by 50%) should be broken down into manageable, engineering-feasible steps (Mercury, Gemini, Apollo).
• Ecosystem Requirement: Achieving these milestones requires a collective ecosystem of academia, startups, and incumbents (e.g., Google, IBM) rather than a single entity.
• Risk Mitigation: By practicing on smaller, predictable "bits" (like docking in space), the industry reduces the risk of catastrophic failure in the final goal.

Forward-Looking Vision for Biological Materials:

• Sustainable Infrastructure: Future engineering may replace steel and metal with living biological materials (e.g., glowing trees that fight global warming).
• Functionality: These biological solutions offer unique advantages in sustainability, growth, and shipping efficiency compared to traditional industrial manufacturing.