Staff Research Engineer (AI)

TL;DR

Staff Research Engineer (AI): Developing next-generation voice-to-voice and multimodal interactive models with an accent on low-latency synthesis and natural conversational flow. Focus on designing multi-modal system architectures and implementing end-to-end generative pipelines from pretraining to production.

Location: Remote (Europe)

Company

Synthesia is the world’s leading AI video platform for business, helping enterprises enhance visual communication through human-interactive models.

What you will do

• Shape the roadmap for new model capabilities and propose novel multi-modal system architectures (text and voice).
• Develop and evaluate streaming and conversational systems for low-latency, interactive voice-video synthesis.
• Implement generative pipelines from pretraining through post-training, utilizing DPO, fine-tuning, and distillation.
• Integrate novel architectures including neural codecs, diffusion, and flow-matching to enhance realism.
• Define new evaluation metrics for conversational systems and curate specialized datasets.
• Ship optimized models to production and iterate based on direct customer feedback.

Requirements

• Strong understanding of generative modeling applied to sequential or multimodal data.
• Hands-on experience with LLMs or transformer-based architectures.
• High proficiency in PyTorch, including distributed training and model optimization.
• Solid grasp of time-series modeling and tokenization in audio, speech, or video contexts.
• Proven experience training deep learning models end-to-end, from data preparation to evaluation.
• Must be based in Europe.

Nice to have

• Experience shipping generative models into live products used at a meaningful scale.
• Background in interactive systems where latency and responsiveness were primary constraints.
• Experience with large-scale LLM training resulting in strong reasoning capabilities.
• Original research contributions in top-tier venues such as NeurIPS, CVPR, ICML, ICLR, or Interspeech.
• Familiarity with SOTA audio/speech architectures like flow-matching or autoregressive decoders.

Culture & Benefits

• Opportunity to work at a high-growth AI company with a $4 billion valuation and premier VC backing.
• Collaborate within a 40+ person R&D department on cutting-edge interactive AI.
• High level of ownership over critical components and the ability to drive broad research directions.
• Remote-first work environment across Europe.