About this role.
Synthesia is seeking a Staff Research Engineer to advance avatar-centric interactive video diffusion models for enterprise AI video products. The role combines applied machine-learning research with production-minded engineering, including multimodal conditioning, real-time long-video generation, perceptual audio understanding, and visual-quality improvements. The engineer will own substantial portions of the model-development lifecycle, from data requirements and experimentation through training, evaluation, and product-ready implementation. Success requires deep expertise in video diffusion, computer vision, PyTorch, and clear communication of research hypotheses and findings. The position is available in a hybrid arrangement at several European hubs or remotely within Europe.
Role DNA
A quick view of the complexity, pace, ownership and collaboration implied by the job description.
Job Complexity
5/5Pace & Pressure
5/5Autonomy Level
5/5Communication Load
4/5Salary analysis
Estimated compensation compared with the broader US market for similar roles.
Core skills
Skills and capabilities most closely associated with this opportunity.
Sample interview questions
I would begin with a modular conditioning design that encodes each signal into aligned temporal representations and fuses them through cross-attention or adaptive conditioning layers. I would train with synchronized multimodal clips, use classifier-free guidance or condition dropout for robustness, and evaluate both per-frame quality and long-horizon temporal consistency. Ablations would identify which conditioning pathways improve lip synchronization, responsiveness, and motion realism.
I would explore a streaming architecture that generates overlapping temporal chunks while carrying forward compact motion, identity, and scene-state representations. Latent-space diffusion, temporal caching, distillation, efficient samplers, and reduced-step inference would be central to meeting latency targets. I would measure throughput, end-to-end latency, boundary artifacts, identity drift, and quality degradation across increasingly long sequences.
I would combine automated, human, and production-oriented evaluation. Automated metrics could cover lip-sync alignment, temporal stability, identity preservation, perceptual quality, motion realism, and response latency; targeted test sets should include varied languages, accents, lighting, poses, and interaction scenarios. Human preference studies and failure-taxonomy reviews would validate whether metric improvements correspond to better user experience, while dashboards would continuously track regressions.
I define explicit hypotheses, baselines, success metrics, and time-boxed experiments before investing heavily in an approach. Promising work is converted into reproducible pipelines with versioned data, configurations, checkpoints, and evaluation reports. I communicate trade-offs early and prioritize changes that improve user-facing quality, inference cost, robustness, or maintainability rather than optimizing benchmark metrics in isolation.
I start by translating product behavior into measurable model requirements and a data specification, including coverage gaps, annotation quality, consent, and safety constraints. With data partners, I establish dataset acceptance criteria and audit slices that reflect anticipated failure modes. With engineering partners, I align early on serving constraints, interfaces, observability, and rollout plans so research prototypes can mature into monitored, maintainable product systems.
Synthesia is the world’s leading AI video platform for business, used by over 90% of the Fortune 100. Founded in 2017, the company is headquartered in London, with offices and teams across Europe and the US.
As AI continues to shape the way we live and work, Synthesia develops products to enhance visual communication and enterprise skill development, helping people work better and stay at the center of successful organizations.
Following our recent Series E funding round, where we raised $200 million, our valuation stands at $4 billion. Our total funding exceeds $530 million from premier investors including Accel, NVentures (Nvidia’s VC arm), Kleiner Perkins, GV, and Evantic Capital, alongside the founders and operators of Stripe, Datadog, Miro, and Webflow.
About the role
As a Staff Research Engineer, you will join a team of 40+ Researchers and Engineers within the R&D Department working on cutting edge challenges in the Generative AI space, with a focus on avatar-centric interactive video diffusion models. Within the team you’ll have the opportunity to work on the applied side of our research efforts and directly impact our solutions that are used worldwide by over 60,000 businesses.
This is a unique opportunity for experts in machine learning and diffusion models to shape the future of AI video agents that can think, act, and react like humans. As part of our Interactive Avatars Team, you’ll work on cutting-edge research with a clear focus on turning breakthrough ideas into real product capabilities. You’ll join a team that moves fast, iterates often, and builds models that ship and make a meaningful impact. Example tasks and responsibilities include:
Adapt diffusion models to incorporate diverse conditioning signals (e.g., audio, motion, interaction cues).
Develop methods for streaming infinitely long video sequences at real-time rates.
Work on the perceptual layer of interactive agents, including understanding user audio and generating appropriate contextual reactions.
Improve lip-sync accuracy, motion realism, and overall visual quality in video diffusion models.
Build robust evaluation frameworks and test suites to enable continuous quality tracking.
Collaborate closely with our data team to define data needs and ensure high-quality datasets.
Stay up to date with research in world models, interactive human/agent modeling, diffusion models, and related areas.
What we’re looking for:
Comfortable owning and executing on the responsibilities listed above.
Strong ML (e.g., diffusion, GANs, VAEs) and computer vision background with relevant industry experience.
Hands-on experience with diffusion models (ideally avatar-centric or video-focused) and up to date with recent advances.
Proficient in PyTorch and familiar with modern ML frameworks and tooling.
Strong Python engineering skills, confident with git and version control, and a commitment to clean, maintainable research code.
Outcome-driven, detail-oriented, and motivated to push state-of-the-art research into real product impact.
Clear communicator of hypotheses, experiments, and results.
What will make you stand out:
Experience with audio-conditioned video diffusion models and deep knowledge of recent video DiT architectures.
Demonstrated ability to own the full model development pipeline end to end, from data preparation to model design, training, and evaluation.
A strong publication record in areas such as world models, interactive agents, or video diffusion models.
Why join us?
We’re living the golden age of AI. The next decade will yield the next iconic companies, and we dare to say we have what it takes to become one. Here’s why,
Our culture
At Synthesia we’re passionate about building, not talking, planning or politicising. We strive to hire the smartest, kindest and most unrelenting people and let them do their best work without distractions. Our work principles serve as our charter for how we make decisions, give feedback and structure our work to empower everyone to go as fast as possible. You can find out more about these principles here.
Serving 50,000+ customers (and 50% of the Fortune 500)
We’re trusted by leading brands such as Heineken, Zoom, Xerox, McDonald’s and more. Read stories from happy customers and what 1,200+ people say on G2.
Proprietary AI technology
Since 2017, we’ve been pioneering advancements in Generative AI. Our AI technology is built in-house, by a team of world-class AI researchers and engineers. Learn more about our AI Research Lab and the team behind.
AI Safety, Ethics and Security
AI safety, ethics, and security are fundamental to our mission. While the full scope of Artificial Intelligence’s impact on our society is still unfolding, our position is clear: People first. Always. Learn more about our commitments to AI Ethics, Safety & Security.
The good stuff…
Competitive compensation (salary + stock options + bonus)
Hybrid work setting with an office in London, Amsterdam, Zurich, Munich, or remote in Europe.
25 days of annual leave + public holidays
Great company culture with the option to join regular planning and socials at our hubs
+ other benefits depending on your location
You can see more about Who we are and How we work here: https://www.synthesia.io/careers
LI-MD1
Annual salary information is not provided for this position. Explore salary ranges for similar roles in our Salary Directory ›
Upload your resume now
- To unlock remote work opportunities and be discovered by global employers.
This job listing has been manually reviewed by the Jobicy Trust & Safety Team for compliance with our posting guidelines, including verification of the company's legitimacy, accuracy of job details, clarity of remote work policy, and absence of misleading or fraudulent content.
