Catalio Conversations

Catalio Capital Management
  • 0.0 (0)
  • Entrepreneurship

Welcome to Catalio Conversations! We spotlight founders & CEOs across innovative healthcare, AI & life sciences, who are building the next generation of disruptors in the space – from drug discovery and autonomous labs to clinical care & beyond. The future starts here.

Episodes

  1. 2 days ago

    How Robotics is Rebuilding the Wet Lab w/ Mostafa ElSayed Co-Founder & CEO of Automata

    What if the biggest constraint on AI for biology isn't the models or the algorithms, but the physical labs that have to generate the data in the first place? Martha Petrocheilos is joined on Catalio Conversations by Mostafa ElSayed, co-founder and CEO of Automata, whose view is that lab automation only looks like a robotics business. Underneath, it's about orchestrating data. The wet lab is being asked to shift from producing experiments to producing information, and the capital pouring into models far outpaces what's going into the infrastructure that feeds them. Infrastructure has historically come before every software wave. Biology is still waiting for its version. Mostafa came to this from architecture and a stint at Zaha Hadid, started Automata in 2016 as a general robotics company whose largest customer was Fiat Chrysler, and only later concluded the lab was where the technology mattered most. His early mistake was diagnosing the field as a robotics failure. Warehousing has real robotics problems. Labs have a heterogeneity problem, and automation has always performed badly where variability is high and volume is low. Automata's response is Link, a modular arrangement of benches and robots that flattens the differences between a hospital lab, a genomics facility and a global pharma site, paired with software that designs the run, executes it, and captures data throughout instead of only at the end. Deployments usually involve several robots and several experiments at once, generating gigabytes to terabytes a week. Mostafa cites a UK cancer diagnostics lab where cost per test dropped from roughly £21 to £2 and turnaround compressed from 25 to 30 days down to eight. He has seen equivalent testing elsewhere in Europe take 150 days. His father died of cancer in 2022, 270 days after diagnosis, which is why those numbers aren't abstract to him. Perfect for healthcare leaders, biotech investors, and anyone weighing where value sits in AI for science, this episode makes the case that self-driving labs arrive narrow and specific first, and that the durable position is the layer everything else has to run on. Timestamped summary: 0:00 — Introduction to Automata 1:09 — From architecture to lab automation 3:55 — The shift from robotics to data orchestration 6:35 — How Automata’s platform works 8:51 — Why lab automation has lagged behind 10:25 — Reducing testing costs and turnaround times 11:32 — Building the connective layer for labs 13:41 — Helping scientists embrace automation 15:54 — AI, infrastructure, and better biological data 17:35 — The lab of the future 19:44 — How close are self-driving labs? 22:22 — Quick-fire questions 22:56 — Closing thoughts and outro

  2. 25 Aug

    How Testing Drugs in Human Cells Cracks Undruggable Targets w/ Sri Kosuri Co-Founder & CEO of Octant

    What if the drugs the industry calls undruggable are simply the ones you can't design on a computer or purify into a test tube, and can only find by running chemistry inside living human cells, hundreds of thousands of times? In this episode of Catalio Conversations, Martha Petrocheilos sits with Sri Kosuri, co-founder and CEO of Octant, who argues AI isn't the bottleneck in drug discovery, data is, and almost nobody will build the machine that produces it. Value accrues not to foundation model companies but to the ones running the discovery loop themselves. The goal isn't the AI drug. It's the AI drug hunter. Sri's path runs from an academic lab at UCLA, where his group made gene synthesis far cheaper, through a DNA synthesis company that was later acquired, to the conclusion that the only way to know was to build the drugs himself. Octant works on correctors, small molecules that rescue proteins the body builds incorrectly — and their difficulty is the point. An unfolded protein has no structure, so structure-based design doesn't apply. It degrades, so it can't be purified. The chemistry has to run in cells, across enormous numbers of iterations. Vertex spent 10 to 15 years and roughly 75,000 compounds to do this in cystic fibrosis, and still has no competition. Octant built 250,000 analogs in one year for its lead retinitis pigmentosa program and now runs 100,000 to 200,000 compounds a month, with a second program in Fabry disease and p53 correctors in cancer behind it. Bristol Myers Squibb invested early alongside Catalio and now has a discovery collaboration. Perfect for healthcare leaders, biotech investors, and anyone trying to separate real AI advantage from the pitch, this episode argues that target-based discovery is becoming commoditized, and the durable business is building drugs no one else can build. Timestamped Summary: 00:00 — Introduction to Sri Kosuri and Octant, a clinical-stage company using synthetic biology, high-throughput chemistry, & AI to build small molecule drugs 00:52 — How a discovery in his UCLA lab convinced Sri to leave academia and start a company 01:56 — Why he chose to develop medicines himself rather than sell tools to other scientists 03:22 — Octant narrows to correctors, drugs that rescue proteins the body builds incorrectly 05:25 — Vertex's cystic fibrosis franchise shows why correctors are so hard to make 06:14 — Early missteps, and how focusing on a single mechanism let those lessons compound across programs 07:00 — Pressure to keep the platform broad, and the partnerships that now cover its carrying costs 08:10 — Inside the platform: building every mutation that might exist in patients & reading their activity in cells 10:11 — Pan-variant correctors and the scale behind them, 250,000 analogs in a year and 100,000 to 200,000 compounds a month today 11:24 — For comparison, Vertex used hundreds of chemists and roughly 75,000 compounds over a decade 12:07 — Direct-to-biology chemistry, multiplexed cell assays, and the engineering needed to run it every week 13:22 — Why an unfolded protein can only be studied in living cells, not through structure or biochemistry 14:53 — Optimizing beyond on-target activity 15:45 — Open AdMet and running blinded competitions on large safety and toxicity datasets 17:07 — Solving off-target toxicity would make small molecules as predictable as antibodies 17:53 — How the Bristol Myers Squibb relationship began, and what pharma partners need to see before betting 19:29 — Data matters more than algorithms, and value accrues to the discovery loop rather than foundation models 21:42 — What changes for patients 23:00 — Quick-fire close

  3. 12 Aug

    How Inhaled Medicine is Transforming Treatment for Rare Lung Disease w/ Lyn Baranowski CEO of Avalyn

    What if the biggest problem in pulmonary fibrosis isn't finding a drug that works — but getting patients to stay on the ones we already have? In this episode of Catalio Conversations, Martha Petrocheilos sits with Lyn Baranowski, CEO of Avalyn Pharma, who makes a contrarian case: two effective medicines have been approved for a decade, yet fewer than 10% of diagnosed patients are still taking them a year later — because the oral formulations are so poorly tolerated that patients simply stop. In a disease with three-to-five-year survival, worse than most cancers, the rate limiter isn't the science. It's the delivery. Lyn's path — from business development at Novartis, through leadership at Pearl before its billion-dollar acquisition by AstraZeneca, to COO at Altavant — is two decades spent almost entirely in respiratory, and it shapes how she reads this market. Avalyn's approach takes pirfenidone and nintedanib and reformulates them for inhalation, putting the drug directly into the lung and cutting the dose dramatically: pirfenidone drops from 2,400 mg a day orally to 200 mg inhaled. Lead program AP01 has now been studied in more than 150 patients, some on therapy beyond five years, with imaging showing stabilization or reversal of fibrosis in 70% of patients on the high dose. Phase 2b data reads out in the second half of 2027, and the company recently went public on NASDAQ in one of the year's largest biotech IPOs, raising roughly $345 million. Perfect for healthcare leaders, biotech investors, and anyone curious about how drugs actually reach patients, this episode makes the case that reformulation is not a tweak but a bet — and that solving tolerability is what finally unlocks combination therapy in a disease that has never been able to use it. Timestamped Summary: 00:07 — Introduction to Lyn Baranowski and Avalyn Pharma, which develops inhaled therapies for rare lung diseases 01:26 — Lynn explains how early respiratory work at Novartis introduced her to pulmonary fibrosis and its unmet need 02:31 — Lessons from asthma and COPD point toward combination treatment in pulmonary fibrosis 03:37 — What drew her to Avalyn: delivering established medicines straight to the lung to keep patients on therapy 04:14 — A patient advisory council reviews protocols and keeps patient needs at the center of strategy 05:19 — Existing oral therapies cause side effects severe enough that most patients stop taking them 05:53 — Avalyn reformulates proven drugs for inhalation, raising lung exposure while cutting the dose 07:02 — Day-to-day treatment is a soft nebulized mist breathed in over about eight minutes 08:38 — Hiring people who developed the original oral drugs gave the team rare disease-specific expertise 09:11 — Fewer than 10% of patients remain on oral therapy at a year, while some Avalyn patients have stayed on beyond five 10:25 — The longer-term goal is making combination treatment practical, as it already is in other lung diseases 11:02 — Clinical data show stabilized lung function and, on imaging, stabilization or reversal of fibrosis 12:06 — The Phase 2b study aims to prove better tolerability alongside sustained lung-function benefit 13:12 — Inhaled reformulation is technically demanding and hard for generic competitors to replicate 14:56 — Doctors need clinical evidence, while patients intuitively grasp the logic of medicine going straight to the lungs 16:03 — Lyn discusses Avalyn's NASDAQ debut and what investors responded to 18:44 — Beyond pulmonary fibrosis, many rare lung diseases still have nothing approved 19:56 — Imaging suggests pirfenidone may help earlier, before fibrosis fully sets in 20:32 — Quick-fire close: her bet is that inhaled delivery unlocks combination treatment

  4. 22 Jul

    How Automated Sample Prep Fixes Biology's Bottleneck w/ Udayan Umapathi Founder & CEO of Volta Labs

    What if the biggest bottleneck in modern biology isn't sequencing the genome — but everything that has to happen before the sample reaches the sequencer? In this episode of Catalio Conversations, Martha Petrocheilos sits with Udayan Umapathi, Founder & CEO of Volta Labs, who makes a contrarian case: the cost of sequencing has collapsed, yet the cost of preparing samples has barely moved — and this overlooked, still-manual step has become one of the biggest constraints on the future of genomics and AI-enabled biology. Udayan's journey — from the MIT Media Lab, where his research on programmable water droplets was rooted in human-computer interaction rather than biology, to founding Volta — showcases what happens when an engineer rethinks a problem from first principles. The result is Callisto, a platform that automates DNA and RNA sample prep using digital fluidics: tiny droplets moved across a flat surface with electric fields, sound waves, and computer vision, each its own miniature reaction chamber. No pipette tips, no robotic arms — just a new physical logic delivering precision, repeatability, higher yield, and lower cost. And Udayan is building an "app store for the lab," one instrument that shifts from one application to the next! Perfect for healthcare leaders, AI enthusiasts, and anyone curious about the hidden plumbing of modern medicine, this episode reveals the shift redefining how biology gets done — from a manual, inconsistent, expensive process into something automated, precise, and scalable. Because the next breakthrough in AI-driven biology and drug discovery may depend less on a smarter algorithm and more on finally fixing the step everyone forgot. Timestamped Summary: 00:03 — Introduction to Udayan Umapathi and Volta Labs, which automates genomic sample preparation. 01:11 — Udayan explains how his MIT Media Lab work with programmable water droplets revealed a strong use case in biology 02:19 — He describes turning a research demo into a real-world company 02:50 — Building Volta required integrating engineering, design, biology, hardware, and software over many experiments 03:56 — Clinical adoption demanded reliability strong enough for irreplaceable patient samples 05:38 — Udayan shares an unused but promising discovery involving precisely controlled droplet splitting 06:09 — Callisto manipulates tiny droplets on a flat surface to automate DNA sample preparation 06:46 — Sound waves, magnetic fields, and computer vision help mix, track, and process samples 07:16 — The platform converts raw biological material into molecules that sequencing machines can read 07:49 — Its advantages include repeatability, gentler sample handling, better yield, and lower reagent use 08:57 — Sample-prep costs have remained high because labs need to support many sample types, chemistries, and sequencers 10:06 — Volta reduces reagent costs and can shorten assay setup from many months to a few months or less 11:09 — The company’s broader vision is one adaptable platform that can support many lab applications 12:49 — Labs can switch between workflows without major reconfiguration, gaining more flexibility and avoiding vendor lock-in 13:23 — Adoption becomes easier once labs try the system: customers can be producing usable samples within days 13:56 — Reliable, consistent sample preparation is essential for producing the clean data AI-powered biology depends on 15:39 — Cheaper, scalable sample prep could accelerate diagnostics, drug discovery, and multi-modal sequencing workflows 17:17 — Udayan identifies oncology and rare disease as major early beneficiaries

  5. 7 Jul

    How AI is Matching the Right Cancer Drug to the Right Patient w/ Ron Alfa Co-Founder & CEO of Noetik

    What if the reason ~95% of cancer drugs fail in the clinic isn't the drug at all – but the patient they were given to? In this episode of Catalio Conversations, Martha Petrocheilos sits with Ron Alfa, Co-Founder & CEO of Noetik, who reveals a radically contrarian thesis: the industry is great at making drugs, but has never cracked how to match the right drug to the right patient, and that fixing this "translation gap" could rescue therapies once written off as failures. Ron's journey – physician-scientist with an MD/PhD from Stanford, a master's in the history of medicine, and six years at Recursion from seed stage to IPO as Head of Research and acting CSO – showcases turning a career at the frontier of tech-bio into a company built to solve oncology's hardest problem. From Noetik's conviction that the best model system for cancer is the human patient, to its industrial-scale data engine and Perturb-Map platform, to Octo, its "virtual cell" foundation model that integrates a tumor's DNA, proteins, gene activity, and tissue imaging into a single system, this episode pulls back the curtain on what genuinely AI-powered drug discovery looks like. And with GSK licensing Noetik's models for $50M upfront – a shift from selling services to licensing AI as infrastructure, an entirely new asset class – Noetik isn't just promising the future of oncology; they're proving pharma will pay for it. Perfect for healthcare leaders, AI enthusiasts, and anyone curious about the next frontier of cancer treatment, this episode reveals the paradigm shift redefining how cancer drugs are discovered – directly from human tumor biology, and matched to the patients most likely to benefit. Be here for the rare inside look at how AI could finally let oncology let go of the guesswork that's dominated it for far too long – because the next breakthrough cancer drug might already exist, waiting for the right patient to be found. Timestamped Summary: 00:04 - Introduction to Ron Alpha and the mission of Noetic 01:09 - The contrarian thesis: ~95% of cancer drugs fail on the wrong patients, not because they don't work 01:53 - Why enriching the right patient populations turns more therapeutics into successes 02:10 - How the definition of a tumor evolved: tissue, pathology, genomics, and now AI 02:31 - Six years at Recursion: what Ron carried into Noetic and what he left behind 03:42 - Making it concrete: what Noetic does with a tumor sample (the Genus/ASCO work) 04:27 - Measurement versus prediction, and selecting the patients most likely to benefit 04:55 - Why the best model system for cancer is the patient – not a mouse or a dish 06:13 - Building models from human tissue, with architectures that run "what if" simulations 06:49 - Perturb Map: manufacturing data at industrial scale by switching genes on and off 08:52 - Octo, the foundation model uniting protein, DNA, gene expression, and tissue imaging 09:06 - What it means for a model to truly understand a tumor, and how you know it's right 10:39 - The GSK deal: $50M to license Octo, and AI models as a new asset class 12:29 - Earning trust with a cautious, "show me the data" pharma industry 13:38 - Beyond the same dozen targets: surfacing hundreds more, from tumor to approved drug 15:17 - Myths and bets: more success over speed, and drugs discovered from human data 16:00 - Closing remarks

  6. 23 Jun

    How Clinical AI is Saving Lives with Suchi Saria Founder & CEO of Bayesian Health

    Explore the future of healthcare with AI that could help hospitals catch life-threatening conditions before they turn critical! In this episode of Catalio Conversations, Martha Petrocheilos sits with Suchi Saria, founder and CEO of Bayesian Health, who reveals how combining cutting-edge AI with deep clinical integration is rewriting the rules of patient care – think a sepsis monitor that flags danger hours earlier and helps hospitals act before it’s too late. Suchi’s journey from academic researcher at Johns Hopkins, Stanford, and Harvard to healthcare entrepreneur offers a masterclass in turning breakthrough research into real-world impact. From Bayesian’s continuous AI monitor and EMR-integrated clinical intelligence layer to its landmark results in adoption and mortality reduction, this episode pulls back the curtain on what genuinely AI-powered healthcare looks like. And with the Coalition for Health AI helping set standards across the industry, Bayesian isn’t just promising the future – they’re building it. Perfect for healthcare leaders, AI enthusiasts, and anyone curious about the next frontier of medicine, this episode reveals the paradigm shifts redefining what’s possible in hospitals. Be here for the rare inside look into the biggest leap in healthcare AI –the next life-saving intervention might already be identified by an algorithm, waiting to change a patient’s outcome. Timestamped Summary: 00:00 - Introduction to Suchi Saria and the mission of Bayesian Health 01:01 - The personal story behind Suchi’s dedication to life-saving AI 01:30 - Transitioning from academia to building an impactful health tech company 02:16 - Filling the gap between AI research and bedside application 03:47 - Achieving FDA clearance for the first continuous AI sepsis monitor 04:37 - Challenges of integrating AI into busy hospital workflows 05:46 - Deep integration with electronic medical records and real-world deployment 07:12 - Building clinician trust and avoiding alarm fatigue 08:32 - Impact of AI systems on clinical outcomes and survival rates 09:24 - Real-life story demonstrating AI’s potential in sepsis detection 10:44 - The importance of medical AI solutions that truly drive action 11:55 - Expanding AI to other critical conditions beyond sepsis 12:23 - How hospitals prioritize use cases based on impact and cost 13:01 - The role of collaborative standards in AI safety and efficacy 14:13 - Differentiating between AI 'theater' and tools with real outcomes 15:07 - How AI will change the roles of doctors and nurses in future hospitals 16:28 - The hospital of the future: proactive, AI-enabled, less administratively burdened 17:23 - Myth-busting in healthcare AI adoption 17:46 - The potential of AI in managing clinical deterioration more broadly 18:10 - Closing remarks and the importance of trust in AI healthcare solutions

  7. 9 Jun

    How AI Drug Discovery is Supercharging Pace of Innovation with Tom Miller Co-Founder & CEO of Iambic

    Explore the future of drug development with AI that could compress decades of research into just years! In this episode of Catalio Conversations, Martha Petrocheilos sits with Tom Miller, co-founder of Iambic, where he reveals how combining cutting-edge neural models with high-throughput experimentation is rewriting the rules of pharmaceutical innovation – think a cancer drug designed, optimized, and FDA-approved in under two years! Tom's journey from Caltech professor to drug discovery entrepreneur – alongside co-founder Fred Manby – offers a masterclass in turning academic breakthroughs into real-world medicines. From Iambic's game-changing Neuroplexer model that predicts exactly how drugs bind to proteins, to their landmark cancer candidate IM1363, this episode pulls back the curtain on what genuinely AI-powered drug discovery looks like. And with strategic partnerships with Lundbeck, Revolution Medicines, and Takeda already in place, Iambic isn't just promising the future – they're building it. Perfect for life sciences leaders, AI enthusiasts, and anyone curious about the next frontier of medicine, this episode reveals the paradigm shifts redefining what's possible in pharma. Don't miss this front-row seat to the biggest leap in pharmaceutical history – the next blockbuster cancer drug might already be designed by an algorithm, waiting to save your life. Timestamped Summary: 00:00 - Introduction to Catalio Conversations and guest Tom Miller 00:29 - Tom’s journey from Caltech professor to drug innovator 01:28 - Collaboration with co-founder Fred Manby 01:55 - Balancing scientific integrity with entrepreneurial agility 02:21 - Lessons from early Iambic experiments 03:13 - AI prediction and high-throughput experimentation 03:43 - Defining drug targets and designing molecules 04:11 - Neuroplexer in predicting drug-protein binding 05:06 - Multi-property optimization in discovery 06:24 - Rapid development of cancer candidate IM1363 06:49 - AI reducing drug development timelines 07:16 - Industry impact of AI-designed medicines 07:59 - Partnerships with Lundbeck and Revolution Medicines 08:55 - Building credibility through collaborations 09:21 - Genuine AI transformation vs. superficial branding 09:45 - Scaling advantage of AI-powered discovery 10:36 - AI’s potential in biological complexity 11:31 - Future of AI in drug development 12:24 - Iambic’s identity: AI platform or drug company? 13:21 - Future goals for Iambic and industry impact 14:13 - Promising targets like transcription factors 14:42 - Closing remarks and encouragement for innovation

  8. 6 May

    How Autonomous Labs are Revolutionizing Drug Discovery with Michelle Lee Founder & CEO of Medra AI

    Unlock the future of drug discovery with AI-powered robots that could revolutionize science as we know it! In this episode of Catalio Conversations, Martha Petrocheilos sits with Michelle Lee, CEO of Medra AI, where she reveals how combining AI & robotics is creating the world's most advanced autonomous labs – think 100 robots working overnight to accelerate cures for disease! Michelle’s journey from chemical engineering and AI research to leading Medra’s groundbreaking "physical AI scientists" offers a blueprint for turning science fiction into reality. Looking ahead, Michelle envisions a world where entire drug pipelines are fully autonomous, enabling any scientist to create life-changing medicines. Perfect for life sciences leaders, AI enthusiasts, as well as tech-curious or business-curious viewers eager to see what’s next, this episode reveals the seismic shifts that will define the next decade of scientific discovery. Don’t miss this glimpse into the autonomous labs of tomorrow – because the next genetic breakthrough could come from a robot, not a human. Timestamped summary: 0:03 - Introduction to Michelle Lee and Medra 0:31 - Michelle Lee's Journey: Discussing the realization of the need for AI and robotics in biology 1:00 - Alpha Fold Inspiration: The impact of Alpha Fold 2 on Michelle's vision 1:25 - Data Generation with Robotics: How robots can scale up experimentation 1:53 - Cultural Values at Medra: Insights from Michelle's diverse experiences 2:21 - Medra's Interdisciplinary Approach: Combining engineering, AI, and biology 3:13 - Physical AI Scientists: Differentiating from traditional lab automation 4:08 - Challenges in Data Generation: The need for more and better data 5:00 - Robotics in Experimentation: Solving data quantity and quality issues 5:54 - Real Lab Stories: How Medra improved data generation for a client 7:09 - Overcoming Skepticism: Transitioning from industrial to physical AI 8:00 - Early Challenges at Medra: Learning from initial business strategies 9:25 - Partnership Model: Medra's collaborative approach with clients. 10:17 - Recognition by Genentech: Being named a partner in AI strategy. 11:05 - Building Trust: Importance of fast communication and transparency 11:56 - Future of Labs: Vision for autonomous labs and their impact 12:43 - Empowering Scientists: Enabling scientists to start their own companies 13:10 - Quick Fire Questions: Myths and future bets in AI and life sciences 14:05 - Conclusion: Closing remarks

  • 8 Episodes

About

Welcome to Catalio Conversations! We spotlight founders & CEOs across innovative healthcare, AI & life sciences, who are building the next generation of disruptors in the space – from drug discovery and autonomous labs to clinical care & beyond. The future starts here.

Information

  • Creator
    Catalio Capital Management
  • Years Active
    2026
  • Episodes
    8
  • Rating
    Clean
  • Copyright
    © 2026 Catalio Capital Management
  • Provider

You Might Also Like