Software Engineer, Simulation

Wayve; United States; On-site; USD 140,000 - 200,000; Full time

Job summary

Wayve is building embodied intelligence for safe autonomous driving. The Core Simulation team develops end-to-end simulation that mirrors real driving, ensuring reproducibility and reliability across components and data.

You’ll work on production-grade C++ and Python code, integrating driving-stack components and validating results at scale. You’ll collaborate with Simulation, Robotics, Autonomy, Evaluation, Platform and Data teams to clarify requirements and resolve integration issues, driving

Qualifications

• Professional software development in a collaborative engineering environment.
• Strong C++ and Python skills with depth in at least one language.
• Experience integrating and debugging components in a complex software system.
• Experience in simulation (robotics or AV) or another complex systems domain.
• Experience integrating ML components (e.g. PyTorch models, CUDA).
• Strong fundamentals: testing, observability, maintainability.

Responsibilities

• Develop production-quality components in C++ and Python.
• Integrate driving-stack components and simulated robot systems.
• Test different simulation configurations, focusing on high-scale open-loop evaluation.
• Investigate differences between on-road and simulated execution across data and interfaces.
• Improve reproducibility, reliability, debugging and tooling.
• Profile workloads and contribute performance improvements.

Skills

C++

Python

Software engineering

Profiling

ML integration

Tools

PyTorch

CUDA

Linux

Job description

Before the detail, here's the challenge you'd help us solve.

We build the embodied intelligence that moves real vehicles safely, and the ecosystem a billion machines will run on in the future. Very few people in AI can say this. Every role here, whatever the team, plugs into that.

About our Simulation Teams

Simulation is a mission-critical part of how we build Wayve's driving intelligence. It lets teams evaluate changes faster, investigate failures and build confidence in the driving stack before it reaches the road.

Wayve's end-to-end driving stack creates a distinctive simulation challenge. There are no intermediate stages where simulated signals can be injected, so simulation has to be end to end too: reproducing the behaviour of a learned driving system across sensor data, model inference, routing, control and vehicle dynamics. Our simulator needs to be realistic enough to give meaningful signal, reproducible enough for teams to trust the results, and descriptive enough to explain the differences between simulated and on-road behaviour.

You’ll join the Core Simulation team. Our open-loop and closed-loop workflows share one codebase, with different components configured depending on the use case. You’ll start with the team focused on high-scale open-loop evaluation, while contributing to shared components that also underpin GAIA-powered closed-loop simulation, where generative world models are integrated with the driving stack. Your role is to enable and productionise these capabilities, not primarily to train the models.

Your day-to-day

• Work hands‑on in our shared, modern C++ and Python simulation codebase
• Implement scoped features and integrate driving‑stack components, simulated robot systems and new vehicle‑platform capabilities
• Support and test different simulation configurations, with an initial focus on high‑scale open-loop evaluation
• Investigate differences between on‑road and simulated execution, tracing issues across component interfaces, data flows, timing and configuration
• Improve automated testing, observability, reproducibility, reliability and developer tooling
• Profile simulator workloads and contribute performance and scalability improvements
• Collaborate with Simulation, Robotics, Autonomy, Evaluation, Platform and Data teams to clarify requirements and resolve integration issues

What you'll be working on

• Developing and maintaining production-quality components in Wayve's simulation system using modern C++ and Python
• Integrating driving‑stack components, simulated robot systems and new vehicle platforms into the shared simulation environment
• Making sure components interact correctly across the simulation workflows they support
• Finding the root cause of on‑road vs. simulation differences across data, inference and simulated components
• Making simulation more reproducible, reliable and debuggable, and faster, so we can run ever larger evaluation workloads efficiently
• Contributing to technical designs and helping evolve the quality and maintainability of the codebase

You should apply if

• You have professional experience developing, delivering and maintaining production software in a collaborative engineering environment
• You have strong software development skills in C++ and Python, with depth in at least one and the ability to work effectively in the other (equal expertise in both isn't required)
• You have experience integrating and debugging components in a complex software system, reasoning across interfaces, data flows, timing and failure modes
• You have relevant experience in simulation (robotics or AV), ideally at the integration level, or in another complex systems domain with clearly transferable challenges
• You have experience integrating and working with ML components (e.g. PyTorch models, CUDA, QNN)
• You have strong software engineering fundamentals: testing, maintainability, code review, observability and pragmatic design
• You understand software performance issues and use profiling or measurement to guide improvements
• You use GenAI development tools (coding assistants or agents) to boost productivity, while carefully reviewing, testing and validating their output
• You communicate and collaborate well, deliver scoped work independently and raise blockers early

Nice to have:

• Experience with middleware or integrating components across complex software systems
• Experience with sensor data such as camera, radar, lidar or GNSS, including modelling uncertainty or noise
• Experience with performance profiling, observability or debugging distributed systems
• Familiarity with large-scale batch processing, cloud infrastructure or GPU-based workloads
• Exposure to world models, NeRF, Gaussian splatting or learned graphics

Not ticking every box? That's totally okay! If you're passionate about autonomy and keen to learn, we encourage you to apply even if you don't meet every requirement.