Location: Toronto / Ottawa
About Dominion Dynamics
Dominion Dynamics is a Canadian defence technology company building a sovereign Autonomous Combat Platform for RCAF and NORAD missions. Our autonomy and C2 layer, Aura Net, is a Canadian alternative to closed foreign stacks, purpose-built for the Arctic operating environment and near-peer threat. Our digital proving ground, Aura Sim, lets us design, train, and validate autonomous tactical behaviour before it ever reaches a test platform.
We are moving autonomy from architecture and CONOPS into deployed, tested capability — and we’re looking for an engineer who can carry a tactical behaviour all the way from concept to a running policy on a test aircraft.
The Role
You will develop the autonomy that flies the platform. Working directly with mission systems, tactics, simulation, and test stakeholders, you’ll turn operational intent into executable behaviour using a hybrid approach: behaviour trees for structured, auditable, doctrine-anchored tactics, and learned policies (RL/IL) where adaptive, high-dimensional decision-making wins. You’ll then prove that behaviour in simulation and help roll it out onto test platforms.
This is a build-and-deploy role, not a research-only role. Success looks like tactical behaviours that are specified with operators, validated in simulation, and demonstrated on hardware.
What You’ll Do
Elicit and codify tactical behaviour. Work with pilots, tacticians, and capability stakeholders to translate combat employment into clear, testable autonomy requirements.
Design hybrid autonomy. Architect and implement behaviour trees for structured tactical logic and integrate learned policies where they outperform, deciding deliberately where each belongs and how they hand off. Map behaviours against our PA/MA/CA autonomy taxonomy and Capability Register.
Develop teaming behaviour. Build multi-agent coordination, target engagement, and survivability across cooperating platforms.
Build in simulation. Stand up and instrument scenarios in Aura Sim to develop, train, and stress behaviours against representative threat and mission sets across our operating environment.
Train and validate learned policies. Work within our RL training pipeline — reward shaping, curriculum design, and large-scale rollouts — and validate policy behaviour against doctrine and mission requirements.
Verify and validate. Test behaviours through the full V&V lifecycle, including safety and constraint enforcement.
Deploy and test on hardware. Take behaviours from sim to test platforms: integration with the runtime autonomy stack, HIL/SIL testing, flight-test support, and closing the loop on observed vs. simulated performance.
Analyse mission data. Use mission logs and performance data to diagnose failures and refine behaviour models.
Collaborate across the team. Work across simulation, sensors/perception, hardware, and flight test to integrate autonomy end-to-end.
What You’ll Bring
Required
Strong software engineering fundamentals (Python and/or C++), with experience building autonomous or decision-making systems.
Hands-on experience with behaviour trees and/or learned policies (reinforcement or imitation learning) — ideally both, and a clear sense of when to use which.
Experience developing and validating autonomy in simulation environments, including scenario design and instrumentation.
Comfort taking systems from simulation to real platforms — integration, testing, and iterating against real-world performance.
Ability to work directly with operational stakeholders and turn tactical intent into engineering requirements.
Sound software practices — version control, code review, CI, and test discipline.
Strongly preferred — the differentiator
Tactical…
(If this job is in fact in your jurisdiction, then you may be using a Proxy or VPN to access this site, and to progress further, you should change your connectivity to another mobile device or PC).