Research Engineer, Post-Training
Deepgram · USA - Remote
About this role
## Research Engineer, Post-Training ### Company Overview Deepgram is the leading platform underpinning the emerging Voice AI economy, offering real-time APIs for **speech-to-text (STT)**, **text-to-speech (TTS)**, and production-grade voice agents at scale. Deepgram’s voice-native foundation models are available via cloud APIs or self-hosted/on-prem software, delivering **high accuracy, low latency, and cost efficiency**. ### Operating Rhythm - **AI-first mindset**: AI use and comfort aren’t optional—they’re core to how teams operate and measure performance. - **Experiment and adapt quickly**: adopt new models and modes, integrate AI into workflows, and continuously push boundaries. - **Move at the pace of AI**: expect rapid change and evolving day-to-day work. ### The Opportunity You’ll help build the engine that solves customer challenges at scale. When customers hit a limit in voice AI, you’ll **measure it, fix it, and ship improvements to every customer in weeks**. **Example projects:** - Expanding models to new languages and regions - Improving TTS pronunciation of names, terms, and numbers - Isolating a speaker’s voice from background noise and crosstalk - Enabling early access to new voice capabilities (e.g., steerable/cloned voices) and using feedback to guide research - Releasing new benchmarks for Voice AI use-cases ### The Role You’ll thrive if you enjoy: - Turning vague customer problems into **clear metrics** - Using whatever works to fix issues (fine-tuning, new data, external models) - Shipping work that customers can use within **weeks** ### The Challenge We’re seeking researchers who: - **Own a customer problem** until it measurably improves in production - Find the **fastest path to a fix**, even with approaches no one has tried - Know when results are **ready for customers** ### What You’ll Do - Identify where models fall short and set improvement targets - Design evaluations to prove whether a model is better—and why - Curate/generate data and fine-tune models - Build working systems quickly and ship them to production ### Requirements / What to Show Us - No specific degree required - Show hard problems you’ve solved in past projects, shipped systems, or open-source work
Listing freshness
CronJobs last confirmed this listing 10h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.