Staff Software Engineer, WellSaid Labs
Machine Learning
More information about location
AI text-to-speech technology
Job no longer available
AI text-to-speech technology
21-100 employees
Job no longer available
More information about location
21-100 employees
Company mission
To empower everyone with AI voice.
Role
Who you are
- To thrive in this role, you have extensive experience with and a solid understanding of ML concepts and best practices, experience with deep learning frameworks (e.g., PyTorch, Tensorflow, etc.), and an understanding of the key elements to a good developer experience.
- You’ve worked on and/or led at least two significant projects that utilized ML to solve a real world customer pain point or create/inform a solution for a customer. Ideally, you also have some combination of the following:
- Has multi-year experience in iteratively and incrementally developing a mature data-driven web service
- Mindset and experience automating delivery pipelines and integrating continuous monitoring
- Affinity for creating modular, scalable, secure, and well-tested code
- Experience with and passion for building reusable tools to increase workflow efficiencies
- A background using data structures and algorithms to process large amounts of data
- Strong cross-team collaboration skills
- Experience using PyTorch to build a deep learning model
- Working experience with most other elements of our tech stack which includes: C++, Python, CUDA, Docker, Google Kubernetes Engine
- An agile, iterative mentality and approach to problem-solving
- (Bonus) Experience with web technologies like HTML, CSS, React, Node
- (Bonus) Experience profiling and optimizing deep neural network performance
- Great attention to detail
- To join our team you must also:
- Be a U.S. Citizen or Permanent Resident
- Pass a pre-employment background check
What the job involves
- As Staff Software Engineer, Machine Learning, you will work closely with the CTO to elevate the developer experience for our state-of-the-art Deep Learning TTS service. This includes optimizing our training pipeline, automating our delivery pipelines, and rewriting abstraction layers
- In your day-today, you'll work on:
- Iteratively and incrementally improving the ease of use, reliability, and performance of our TTS service
- Building tooling for automating custom voice creation
- Setting up a CI / CD pipeline
- Developing data structures for loading, processing, and validating 1000s of hours of text and speech data
- Improving the performance and the scalability of our machine learning training pipeline by integrating tools like DeepSpeed, PyTorch AMP, PyTorch JIT, and PyTorch Profiler
- Measuring and monitoring our TTS deployment including managing standard metrics like latency, performance, and cost as well as keeping track of TTS metrics like loudness, cadence, and naturalness
- Improving the scalability of our TTS deployment
- Being mindful and conscientious of AI ethics and the social impact of the work you do at WellSaid Labs including advocating for what you believe is right
- Open-sourcing reusable libraries and tooling
Our take
WellSaid Labs offers creative teams a real time text-to-speech engine that leverages AI to produce natural-sounding voice overs and clips, all based on the voices of real people.
The most common rebuttal to this kind of technology is how it can be used with malicious intent – deepfakes being the most pertinent concern. This is precisely why WellSaid Labs has no current intention to bring its product to a consumer market – a wise decision. It’s also worth noting that all its voice samples are provided with consent of the original talent.
Provided the pricing point is right, there’s an obvious attraction to being able to quickly create audio content at an enterprise level. And the fact that the startups first major funding round was oversubscribed indicates that investors see this attraction too. It'll be an interesting company to keep an eye on in coming years.
Steph
Company Specialist at Welcome to the Jungle
Insights
67% employee growth in 12 months
Company
Funding (1 round)
Jul 2021
$10m
SERIES A
Total funding: $10m
Company HQ
Northlake, Seattle, WA
Leadership
Michael Petrochuk
(Founder & CTO)Previously founded AI fashion advice app Wardrobe.Tips and location based social network Chirpp. Former Developer and Engineer at vCita and Stylize, also interning at Google and Uber.
Brian Cook
(CEO)Founder of Incredible Capital. Previously co-founded and served as the CEO at Hyperfish, Nintex, and OBS.
Share this job