Recorded live at New York Tech Week, Karl and Erum sit down with Brenton Alexander (CTO at Roebling) to unpack one of the biggest bottlenecks in scaling “biology as technology”: figuring out what it really takes to design and finance physical infrastructure.
Brenton walks through how Roebling uses AI alongside deterministic engineering models (physics/thermodynamics) to accelerate early facility design, generate capex/opex estimates with uncertainty ranges (not false precision), and help teams run scenarios fast—so founders, investors, and operators can make better go/no-go decisions earlier, reduce wasteful iteration across siloed teams, and focus human expertise where it matters most.
Watch the video
Grow Everything brings the bioeconomy to life. Hosts Karl Schmieder and Erum Azeez Khan share stories and interview the leaders and influencers changing the world by growing everything. Biology is the oldest technology. And it can be engineered. What are we growing?
Learn more at www.messaginglab.com/groweverything
Chapters
(00:00:00) Welcome to Grow Everything Live at NY Tech Week
(00:02:10) The “infrastructure gap”: why feasibility work is slow and expensive
(00:03:05) What Roebling does: accelerating the path from R&D to final investment decision
(00:05:05) Live demo setup: building a yeast-based fermentation facility for a red bio-dye
(00:07:15) What the platform decides (and why inputs matter): equipment, DSP, and cost drivers
(00:10:00) “Why not just use Claude?” Deterministic models + AI tooling for defensible results
(00:14:30) Handling uncertainty: ranges, distributions, and Monte Carlo-style scenario runs
(00:18:40) What changes for engineers/consultants: shifting effort from manual work to judgment
(00:23:10) Reading the outputs: capex/opex, IRR, and the “tornado chart” of uncertainty drivers
(00:28:10) Audience Q&A: logistics/customer delivery, AI’s impact on costs, review fatigue, and assumptions
(00:29:30) Long-term direction: more fidelity, narrower bounds, EPC-ready handoff
(00:30:05) Audience Q&A begins
(00:30:30) Q1: logistics + customer delivery costs (not just “at the gate”)
(00:32:55) Q2: how AI changes operating cost assumptions over time
(00:34:15) Q3: review fatigue—how to structure checks and triage what matters
(00:36:10) Q4: what did the model assume for “colorant”? (and why specificity matters)
(00:38:15) Wrap-up + thank-yous
