Dominik Tornow is the co-founder and chief executive of Resonate HQ and author of Think Distributed Systems. He develops programming models that help distributed applications and AI agents survive crashes, delays, and inconsistent information without burying developers in infrastructure complexity.
Tornow studied software systems engineering at Germany’s Hasso Plattner Institute and worked at SAP, Cisco, and Temporal, including principal-engineering roles at Cisco and Temporal. Before founding Resonate, he developed a particular interest in conceptual and formal models that define a system’s guarantees independently of its implementation.
In June 2023, Tornow founded Resonate with longtime colleague David Farr; the company announced $3 million in seed funding in June 2025. Its central programming model, distributed async/await, extends familiar asynchronous code across machines and time: operations can pause, recover from interruptions, and resume without requiring developers to assemble their own orchestration machinery. His minimal distributed async/await framework explores the same ideas directly.
Tornow began publicly outlining an illustrated distributed-systems book in January 2023 and announced its Manning publication agreement the following year. Published in August 2025, Think Distributed Systems examines failures, transactions, replication, consensus, and the challenge of producing reliable behavior from components with incomplete local knowledge.
- Durable execution without infrastructure lock-in. Tornow favors durable functions and promises that preserve application state across failures while operating on infrastructure teams already use. He opposes mandatory proprietary queues and demonstrates the approach with an AI reminder assistant capable of suspending and resuming long-running work.
- Deterministic simulation testing. His seeded simulation approach makes crashes, unusual event orderings, and stale reads reproducible. Simulation traces can reveal information unavailable to production code, helping engineers and coding agents understand precisely why an algorithm violated its guarantees.
- The specification is the product. Tornow argues that coding agents can translate one abstract protocol into implementations tailored to infrastructure such as Postgres or NATS. His AI Engineer World’s Fair presentation outlines a specific sequence: abstract specification, simulated implementation, concrete specification, then production code. The intermediate simulation enables agents to help design distributed algorithms, including behavior under concurrency and partial failure, instead of merely generating plausible implementations.
His Systems Engineering for Agentic Applications project applies these distributed-systems principles to autonomous software intended for continuous production operation.
