Technical Lead Manager, Synthetic Data

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.

Here’s what this particular role covers.

🛠️ About our Simulation Teams

Simulation is advancing our end-to-end autonomous driving research. The team’s mission is to accelerate our journey to AV2.0 by incubating capabilities that become company-level advantages. GAIA, our generative world models, and the synthetic data they produce, are one of those. This role leads Synthetic Data within Simulation. The team exists to dramatically reduce our dependency on expensive and time-consuming on-road data collection by turning generative world models (models like GAIA-3) into a production engine for training-grade experience. When we can restage real driving onto a camera rig that does not exist yet, rewrite ego motion to create scenarios we have never encountered, and land that data in the same training stack we use for real driving, we can train, evaluate and deploy on vehicles and in geographies we have barely collected from. As Tech Lead Manager, you’ll lead a high-performing team of machine learning engineers and help us answer questions like: can we train and validate a driving model for a vehicle platform before the fleet exists, can synthetically generated data replace scarce real-world data for training and evaluation, and how quickly can we deploy autonomous driving in a geography where we’ve never collected AV data?

🧠 Your day-to-day

Algorithmic and system design: define approach to GAIA generation, system design of the synthetic data pipeline and controllability engine, focus on efficiency and quality.

Cross-team collaboration: aligning with driving-model owners and with evaluation and safety team on utility and scalability of generated data.

Mentorship / async review: reviewing a teammate’s experimental design, PR, or ablation write-up and giving direct feedback.

1:1s and feedback: building rapport and managing your direct reports.

Up to date with state of the art: keeping current on video generation, camera transfer, distillation — insofar as it changes a checkpoint or a mix we ship.

Strategy: leading or joining strategic discussions on the synthetic data roadmap, safety case and platform bring-up sequencing.

🧩 What you’ll be working on

Architect the future — set the technical direction for how we post-train and condition world models for synthetic-data capabilities (rig transfer, pose transfer, controllability), holding a high bar for what counts as training-grade generation.

Own the loop end to end — make sure generation, evaluation and training stay one system: from checkpoint and config, through large-scale GPU inference, to artefacts that land in driving-model training with reproducible lineage.

Get hands-on when it matters — lead from the front on key components, codebases and experiments.

Push throughput and yield — drive inference optimisation (distillation, few-step sampling, KV caching, step count), valid-generation rate, and self-serve workflows so model developers can request synthetic sets without a specialist in the loop.

Disrupt thoughtfully — challenge assumptions about where synthetic data pays off, ask sharp questions, and champion bold ideas that move us beyond incremental gains.

Make things happen — lead a high-performing, cross-functional team of ML engineers and applied scientists working across generative modelling, generation infrastructure and training. Drive quarterly planning and execution in a high-ambiguity environment where the target moves.

Align and connect — collaborate with world-model researchers, platform and infra engineers, driving-model owners and evaluation so synthetic data is integrated into the broader stack, not delivered over a wall. Manage upwards and laterally to align your team’s goals with company priorities and OEM programme timelines.

Architect teams — grow and structure a resilient team by hiring top talent, designing effective operating models, and fostering a sense of belonging regardless of location. Cultivate a strong, inclusive culture rooted in scientific rigour, collaboration and curiosity.

Level up — coach and mentor team members, tailoring growth plans to individual strengths and aspirations. Lead by example through technical engagement and clear feedback.

Champion change — navigate your team through evolving research priorities and fast-moving execution, maintaining stability and trust through uncertainty.

🙌 You should apply if

5+ years of experience in ML engineering or applied research roles, with a track record of training and shipping neural networks — not only operating data platforms.

4+ years of people management experience, including direct reports and cross-functional project ownership.

Deep knowledge of generative modelling (diffusion, flow matching, autoregressive, or VAEs) applied to video or other high-dimensional temporal data.

Hands-on experience with video, generative or world models — for example video generation, novel-view synthesis, neural rendering, or controllable generation.

Working knowledge of cameras and 3D geometry (multi-camera rigs, intrinsics/extrinsics, warps and reprojection) and why they break generation or downstream training.

Evidence of closing the loop: taking generated or simulated data into a trained downstream model and measuring impact through mix ratios, ablations and failure analysis.

Experience operating generation or training at real scale — multi-GPU jobs, workflow orchestration, large video artefacts — and making that path reliable.

Strong Python and PyTorch engineering fundamentals, and experience building research-grade production tools.

Excellent communication skills and a passion for coaching and mentoring others.

You balance technical depth with people leadership. You know when to lead from the front and when to empower your team.

You embrace ambiguity and help your team make sense of it, keeping clarity and momentum through uncertainty.

🌱 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.

More about Wayve:

🚀 Wayve is building the leading AI platform for autonomous driving. We are pioneering an end to end AI approach that enables vehicles to learn directly from real world experience, developing the ability to adapt, generalise and improve at scale. Instead of relying on hand coded rules or pre mapped environments, our AI Driver learns to drive by understanding the world around it. The result is technology that navigates complex urban environments with intelligence, precision and natural flow, unlocking meaningful advances in both safety and efficiency. We believe autonomy represents a once in a generation transformation in how people and goods move, comparable to the shift from horses to cars, and from human driven vehicles to intelligent machines.

Our ambition is to make autonomy universal. Wayve’s mapless and hardware agnostic AI platform integrates with global OEM partners, enabling continuous software evolution and unlocking advanced levels of automation from L2 plus through to L4 as our core AI model scales. In a race increasingly defined by intelligence and real world learning, Wayve is taking a distinct approach, building a generalisable driving intelligence that can power any vehicle, anywhere. By combining embodied AI with scalable deployment, we are creating technology that can be shaped to each OEM brand and driver experience, accelerating the transition to a safer, more intelligent future of mobility.

How we work 💻- Locations & Flexible Working:

Our main hubs are in London, Sunnyvale, Yokohama, Herzliya, Vancouver and Leonberg. We o