Jonathan Hurst, co-founder and chief robot officer at Agility Robotics, is a robotics academic-turned-operator focused on deploying humanoid robots in real-world industrial settings. The session, moderated by Ankur Saxena of TDK Ventures, explores how to translate deep tech into scalable businesses. 

  • From demo to deployment
    • Distinguishing real innovation from hype is difficult; simulations and polished demos often mislead. 
    • The true test is real-world deployment: systems must be reliable, cost-effective and repeatable—typically “10–100x harder” than demos. 
    • Strongest early signal: customers willing to buy and expand usage. 
  • Technology-first vs problem-first ventures
    • Agility began with a technology thesis (robots operating in human environments), not a defined use case. 
    • This approach can work for rare, enabling technologies but requires years to find product–market fit. 
    • The company deliberately narrowed focus to a “beachhead” use case (warehouse tote movement) to start a commercial flywheel. 
  • Importance of problem framing
    • Many failures stem from poorly defined problem statements, not weak technology. 
    • Success requires deep understanding of customer needs, application constraints and economic viability. 
    • Knowing what not to pursue is as important as selecting the initial use case. 
  • Avoiding dead ends in innovation
    • Strategic choices should build capabilities that unlock future markets, not narrow, one-off solutions. 
    • Example: investing heavily in safety systems (despite short-term cost) enables broader deployment across industries. 
  • Execution over theory
    • Transitioning from lab to market exposes unknown challenges; there is no substitute for real-world iteration. 
    • Academic approaches often misidentify practical problems; execution requires adaptability and humility. 
  • Investor-relevant success metrics
    • Long-term success depends on aligning technical progress, scale, and economic value.
    • Building a “flywheel” of deployment, learning and revenue is critical.
    • Alignment of incentives (investors, founders, partners) simplifies decision-making and avoids conflicts.
  • Implications for corporates
    • Innovation often fails at two extremes: incrementalism or overly abstract research. 
    • The optimal approach lies in bridging academia and industry—combining rigorous problem framing with real-world constraints. 

This is an AI-generated summary, which has been lightly edited by GCV staff.