Speechmatics is a cutting-edge applied AI Research company that is breaking down cultural barriers by building diverse and inclusive speech technology. We are looking for an experienced Machine Learning Engineer who can help us advance our automatic speech recognition (ASR) engines for Flow, our new Conversational AI companion and create interactive voice interfaces of the future that ‘Understand Every Voice’.
We are betting big on scaling models, so you will be working with millions of hours of audio and billion parameter models which train across dozens of GPUs. It’s all about finding bottlenecks across our 50+ languages.
Our main research focuses are large-scale self-supervised learning, building state-of-the-art speech pipelines and breaking ground on our own emotive text-to-speech (TTS) models.

YOU’LL WANT TO JOIN OUR TEAM IF YOU

• Ambitious Engineers keen to work on bleeding-edge speech recognition and representational learning.
• Experience owning areas of code, seeking alignment rather than guidance.
• Proven-track record of delivering results, moving fast and keeping things simple.
• Someone who loves working in collaborative and diverse teams.
• Has a growth mindset and loves to develop oneself and others.
• Enjoys solving challenging problems and optimising a stack of unfamiliar code.
  We encourage you to apply even if you do not feel you match all of the requirements exactly. The list of requirements is intended to show the kinds of experience and qualities we’re looking for, but it is not exhaustive. If you are interested in the role, the team, and our mission, we would love to consider your application. We are always open to conversations and look forward to hearing from you.

WHO WE ARE:

Speechmatics is the leading expert in Speech Intelligence, and uses AI and Machine Learning to unlock business value in human speech worldwide. We work with an amazing mix of global companies, and our technology can integrate into our customers stack irrespective of their industry or use case – making it the go-to solution to harness useful information from speech.
Joining us means working with some of the smartest minds around the world, focused on cutting-edge projects and deploying the latest techniques to disrupt the market. We believe in putting people first; we’ll do all we can to help you develop your skills and give you the tools you need to thrive. Our Focus Fridays give you an undisturbed day of focus, offset with Together Tuesdays when we have our team meetings, so you’ve always got the right balance.
We have structured a hybrid approach that includes 2-3 designated office days each week. This arrangement ensures that while we embrace the advantages of remote work, we also maintain the vital connection and synergy that only in-person interactions can foster.
This is only the beginning; we’re looking for amazing people like you to continue our journey…

WHAT YOU’LL BE DOING

• Working with a diverse group of engineers across Speechmatics.
• Scaling self-supervised learning models across hundreds of GPUs in the cloud.
• Experimenting with distillation or quantisation to speed up our models at runtime.
• Comparing compute efficiencies of architectures such as a transformer and the impact on WER.
• Developing new, state-of-the-art AI features, from training models, all the way to shipping it to production.
• Advancing end-to-end speech models and researching new approaches to TTS.
  We aim to get you onboarded and started on projects in your first few days. In addition, having a very collaborative culture, you will often be pair programming with a colleague on streamlining our production ML pipelines, reviewing other folks’ code and suggesting new ways to tackle a tough real time factor (RTF) optimisation problem, as well as brainstorming novel approaches for analysing model predictions with the team.