About kea
Launched in 2018, kea is one of the fastest-growing companies changing the way restaurants operate. We've raised $29 million from Tier 1 Silicon Valley investors, and we're not stopping there. Originating in Mountain View, California, and now fully remote, we are building a team that is excited about our mission: drive the world’s commerce by empowering restaurants to operate more intelligently and own their relationship with customers. The first product in our platform, kea Voice, takes orders over the phone at a high level of efficiency and consistency so that restaurants can focus on making delicious food. That's a win-win!
Our Commitment as a team
At kea, we live our values--we are committed to innovation while moving as one unit at high velocity. We are compassionate and care for others. We are open and transparent. We play to win the long game. As you review the below keep Our Commitment in mind. Your uniqueness is valued and differences of opinion, background, and ability are welcomed. If you think you satisfy even just a few of the below points, please apply!
About the Role
You will be a part of a team that is automating the order-taking process. Your mission will be to help our team find the best solution for transcribing audio from phone calls to interpret customer requests. This role plays a critical part in our organization, collaborating closely with the ML Ops member as well as the software and testing engineers to ensure the proper integration of ASR models into our ecosystem.
What your day-to-day looks like
- Upgrading and optimizing our ASR models
- Integrating ASR models with NLU pipelines
- Collaborating with both the ML Ops member as well as the software and testing engineers
- Identifying and solving challenges related to generating high-quality transcriptions of customer utterances
- Identifying and resolving ASR error propagation issues
What sets you up for success
- Basic experience in ML, programming, and acoustic modeling
- A minimum of 3 years of software engineering experience
- A minimum of 3 years of experience with ML
- A minimum of 1 years of experience with ASR
- Bachelor's degree in Computer Science, Data Science, Physics, Mathematics, or an equivalent field
- Knowledge and experience in Machine Learning concepts, NLP, Speech Processing, Acoustic and Language Modeling, Data Collection and Augmentation, Evaluation Metrics, and Software engineering with Python
- Familiarity with ASR frameworks such as Kaldi, Wav2Vec, Whisper
- Experience with data synthesis techniques like speed perturbation and room simulation
Nice-to-haves
- Academic publications in ASR, SLU, NLP, or NLU
- Experience in selecting and deploying ASR models in a production environment
- Familiarity with text-to-speech providers like GCP Speech-to-text, AWS Transcribe, or Deepgram
What we offer:
- Generous stock option plan
- 100% employer-paid medical and mental health care benefits. Vision and dental are also available
- Flexible time off
- Remote work
- 401K
- Ownership, responsibility, and empowerment in what you do
- Incredible teammates and a caring workplace culture