Open Roles

We’re currently hiring for Generative Speech (LLM-based TTS). Other roles are planned and shown below for visibility.

Open

Generative Speech — LLM-Based TTS

Research Engineer / Applied Scientist

Build controllable, identity-preserving speech synthesis for real-world video localization.

LLM-conditioned control (style / emotion / intent)
Speaker identity preservation + long-form stability
PyTorch training / finetuning + eval loop

No formal application — just send what you have.
Optional: LinkedIn, CV, GitHub/paper link, or short intro.

LLM-conditioned TTSvoice identityprosody/style controlmultilingual
Planned

Video Text Spotting — Detection, Tracking & Recognition

Applied Scientist / CV Engineer

Detect and track on-screen text in videos, then recognize it reliably across frames.

Text detection + tracking across time (scene cuts, motion blur)
Recognition with multilingual support
Dataset curation + evaluation for real-world overlays
Join talent pool
Not accepting applications yet
Opening soon

Optional: send a short intro + relevant work.

Planned

Video Motion Typography Translation — Layout-Aware Editing

Applied Scientist / Research Engineer

Translate animated on-screen typography while preserving motion, timing, and design.

Track text regions + motion paths (warp, perspective, occlusion)
Style-consistent rendering (fonts, stroke/shadow, effects)
Frame-accurate timing + seamless compositing
Join talent pool
Not accepting applications yet
Opening soon

Optional: send a short intro + relevant work.

Planned

Agentic LLM — Workflow Automation for Localization

ML Engineer / Applied Scientist

Build tool-using agents that orchestrate translation → terminology → QA → synthesis reliably at scale.

Agent planning + verification
Tool integration (search, QA, evaluation)
Production-grade pipelines & monitoring
Join talent pool
Not accepting applications yet
Opening soon

Optional: send a short intro + relevant work.