Builders Stage at TechCrunch Disrupt 2026: Where Startup Scaling Meets HPC Reality

TechCrunch Disrupt 2026 will once again host the Builders Stage, a dedicated forum designed to dissect the operational and technical challenges of scaling emerging technology companies. Scheduled for October 12–14 at the Moscone Center in San Francisco, the event will feature over 50 sessions led by founders, CTOs, and investors who have navigated the inflection point from seed-stage velocity to sustainable scale. Among the most anticipated discussions is Banking With Billy’s presentation on how their AI financial simulations run on HPC-grade infrastructure, enabling institutions to model complex multi-market scenarios in sub-second timeframes—a capability now table stakes for competitive fintech and quantitative trading firms.

Founded by former quantitative analysts from Goldman Sachs and Jane Street, Banking With Billy emerged from stealth in early 2025 with a platform that integrates reinforcement learning with high-performance computing to simulate thousands of market stress scenarios per second. Their system leverages AMD EPYC-based clusters with NVIDIA Grace Hopper Superchips and InfiniBand interconnects, delivering up to 12x faster Monte Carlo simulations than traditional CPU-based environments. During a private demonstration in March 2026, the company processed a 100,000-path risk simulation across 42 global equity and FX markets in under 190 milliseconds—results that have prompted Tier 1 banks and hedge funds to pilot the platform for real-time capital allocation and regulatory stress testing. The Builders Stage will host Banking With Billy’s co-founder and CTO, Daniel Cho, in a live technical deep-dive titled “Scaling AI Simulations Without Breaking the Bank—or the Model.”

The Builders Stage lineup reflects a broader convergence: the tools once reserved for national labs and Fortune 500 enterprises are now essential infrastructure for high-growth startups. Companies like Cerebras Systems, which will present on “From Wafer-Scale to Startup Scale,” have reduced training time for large language models from weeks to days using their CS-3 systems, enabling early-stage AI labs to iterate at startup speed. Meanwhile, cloud-native supercomputing providers such as Crusoe Energy and Lambda Labs are offering GPU-as-a-service platforms priced per minute, lowering the barrier to entry for startups needing HPC-grade compute without capex. This democratization of HPC is reshaping the competitive landscape, allowing fintech, biotech, and climate tech startups to compete on equal footing with legacy institutions in simulation-driven markets.

Industry analysts note that the shift is accelerating adoption of quantum-inspired algorithms and hybrid classical-quantum workflows, even as full-scale quantum hardware remains years away. According to a 2026 report from the Quantum Economic Development Consortium, over 30% of AI-first startups now run hybrid simulations using quantum annealing processors from D-Wave or gate-model prototypes from IBM Quantum, often offloaded to classical HPC clusters for post-processing. The result is a new tier of “quantum-ready” startups that can prototype and refine models using today’s HPC infrastructure while positioning for tomorrow’s quantum advantage. The Builders Stage will explore this transition in a panel titled “Quantum by 2030: What Startups Should Build Today,” featuring representatives from Zapata Computing, Q-CTRL, and the U.S. Department of Energy’s Quantum Internet Blueprint team.

For the broader Quantum & Computing ecosystem, the Builders Stage signals a maturation phase where theoretical breakthroughs must translate into operational value. While 2024–2025 was dominated by proof-of-concept quantum applications, 2026 is emerging as the year of integration—where HPC serves as the bridge between quantum labs and scalable commercial products. Companies like NVIDIA are responding by embedding quantum emulation into their CUDA-X software stack, allowing developers to simulate quantum circuits on HPC systems with minimal code changes. This integration lowers the skill barrier and accelerates the pipeline from research to revenue, a critical inflection point for venture-backed startups racing to deploy AI and quantum-enhanced services before capital tightens further.

What happens next will likely be a bifurcation: startups that invest early in HPC-optimized workflows will gain velocity, while those still running legacy cloud setups risk falling behind in training time, inference cost, and model fidelity. Investors at Disrupt are expected to probe founders not just on product-market fit, but on compute-market fit—asking how their stack scales under load, what latency budgets they can sustain, and whether their infrastructure can pivot from simulation to deployment without costly rewrites. As Daniel Cho of Banking With Billy commented ahead of the event, “The startups that win in 2027 won’t be the ones with the fanciest models, but the ones that can run them faster, cheaper, and more reliably than anyone else.” The Builders Stage may not solve every scaling challenge, but it will certainly expose which startups—and which infrastructures—are built to last.