How AI Will Accelerate Breakthroughs in Biotechnology with Benchling CEO Sajith Wickramasekara
Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
Podcast Summary
In this fascinating episode of No Priors, Sarah Guo interviews Sajith Wickramasekara, co-founder and CEO of Benchling, about revolutionizing biotech R&D through AI. (05:34) The conversation explores the challenges of bringing new drugs to market, where over $2 billion and 7-10 years are required, with most medicines failing late in development. (10:29) Sajith discusses how the biotech industry is experiencing a "dot com bust" equivalent, making faster, cheaper drug development more critical than ever. (04:25) The discussion covers Benchling's AI agents that help scientists make better decisions and experiment faster, from simulation tools to deep research capabilities that can uncover institutional knowledge trapped in old lab notebooks. (18:00)
• Main Theme: How AI agents and systematic approaches to scientific data can transform biotech R&D from an artisanal, inefficient process into a faster, more predictable system for bringing life-saving medicines to patients.Speakers
Sajith Wickramasekara
Co-founder and CEO of Benchling, the system of record for biotech R&D that serves over 1,300 biotech and pharma companies including Moderna, Sanofi, Eli Lilly, and Regeneron. A software engineer by background who worked in biology labs, Sajith founded Benchling 13 years ago after recognizing the stark contrast between the sophisticated tools available to software developers and the paper notebooks and spreadsheets used by scientists. Under his leadership, Benchling has become the central platform powering innovation across the entire biotech sector, from cutting-edge AI biotechs like Isomorphic Labs to established pharmaceutical giants.
Sarah Guo
Host of No Priors podcast and venture capitalist focused on AI and technology investments. Sarah brings expertise in evaluating AI applications across various industries and has deep knowledge of both the technical capabilities and business implications of artificial intelligence in enterprise settings.
Key Takeaways
Medicines Are Magic With Incredible ROI
Prescription drugs represent only 9% of healthcare spending in the US, yet they provide extraordinary return on investment because they become cheaper over time when they go generic, unlike other healthcare services that become more expensive. (09:54) This insight reframes the conversation around drug costs - while $2 billion seems expensive for drug development, the long-term societal benefits are massive. Statins that cost pennies today were expensive medicines twenty years ago, demonstrating how pharmaceutical innovation creates lasting value that compounds over time.
The Power of Institutional Memory in Science
One of Benchling's most impactful AI applications helps companies avoid repeating expensive experiments by surfacing historical data trapped in old lab notebooks. (17:39) A customer planning 20 different mouse studies worth eight months of work discovered through Benchling's deep research agent that many of these studies had already been conducted years ago by people who had since left the company. This highlights how much valuable scientific knowledge gets lost in "folklore and institutional knowledge" that disappears when researchers leave, making AI-powered data organization and retrieval incredibly valuable for accelerating research.
The Translation Problem in AI for Science
The biggest challenge in AI for biotech isn't the underlying technology but making it accessible and trustworthy for scientists. (22:08) Sajith describes this as having "GPT but no Chat" - the capabilities exist, but the interface that makes them truly useful hasn't been figured out yet. Most scientists outside of Silicon Valley aren't using much AI in R&D yet due to concerns about accuracy, intellectual property, security, and legal issues. Success will come to whoever can make AI tools that scientists actually trust and want to use in their daily workflows.
Biotech Must Become Less Artisanal
The fundamental problem with drug development is that every company reinvents workflows, data structures, and processes from scratch because they're playing a "one-time game" - trying to survive until they can show clinical success and get acquired. (12:22) This artisanal approach prevents the systematization that has made other industries more efficient. Companies don't invest in scalable systems because they don't expect to need them long-term, creating massive inefficiencies across the entire industry. AI offers an opportunity to standardize and systematize these processes, potentially reducing the 7-10 year timeline and $2 billion cost.
Focus on Better Molecules, Not Just Clinical Trials
While clinical trials get attention because they're the biggest line item in drug development, Sajith argues this focus is "a bit of a red herring." (13:32) The real problem is that many molecules simply aren't good enough - they fail not due to operational issues in trials, but because they don't work safely and effectively. The industry needs to focus on creating better molecules and moving them to clinical testing faster and cheaper, rather than just optimizing the clinical trial process itself. This requires better predictive models and simulation tools that can identify promising compounds earlier in the process.
Statistics & Facts
- Over $2 billion and 7-10 years are typically required to bring a new medicine to market, with most medicines failing late in the development process. (10:29) This statistic underscores why biotech is such a difficult industry for investors to underwrite and why systematic improvements are desperately needed.
- Prescription drugs represent only 9% of healthcare spending in the US, yet provide extraordinary value because they become cheaper over time when they go generic. (09:54) This puts pharmaceutical costs in perspective and highlights why we should want more medicines rather than fewer.
- Benchling currently works with about 1,300 biotech and pharma companies and scientists at over 7,000 academic institutions worldwide. (01:39) This scale gives them unique visibility into innovation happening across the entire biotech sector.
