OpsLab’s scheduling platform saves Air Force squadrons 200 hours per month — time previously spent by pilots manually fixing flight schedules until late at night. Arun Nair, Founder & CEO, showcases how OpsLab’s three-layer intelligence system went from 30% scheduler acceptance to operational deployment across 75+ squadrons by combining operations research algorithms, reinforcement learning from daily user feedback, and LLM interfaces that let commanders simulate scenarios like losing 20% of their fleet while managing ATOs and continuation training simultaneously. Arun warns how Phase III SBIR contracts aren’t the finish line everyone thinks they are — the real Holy Grail is program-of-record status. His team spent four years collecting feedback from schedulers to calibrate their algorithms, hired former Air Force schedulers to understand the dysfunction, and built a middleware layer borrowed from his Wall Street high-frequency trading experience that can combine air power from different branches onto the same platform. Topics discussed: - Building AI scheduling architecture combining operations research, reinforcement learning from human feedback, and LLM interfaces. - Achieving 200 hours monthly time savings per squadron by eliminating manual scheduling work that kept pilots working late nights. - Scaling across 75+ squadrons and multiple aircraft types from single-seat fighters to B-52 bombers with eight crew positions. - Understanding Phase III SBIR contracts are still R&D phase, not a sustainable business model, with program of record as the true goal. - Hiring former Air Force schedulers to decode organizational dysfunction and collecting daily feedback to calibrate algorithms. - Targeting 60 VCs for Series A fundraising, narrowing to 15 serious prospects, and using venture debt to extend runway. - Applying Wall Street high-frequency trading middleware concepts to create a pub-sub layer combining air power from multiple military branches. - Deploying LLM-powered scenario simulation enabling commanders to model fleet losses, ATOs, and continuation training constraints simultaneously.