This is a remote position.
Coders Connect has joined forces with an innovative startup a cuttingedge technology company committed to transforming speaker verification. We are currently in search of a skilled Machine Learning Data Scientist with a specific proficiency in Speech Science particularly emphasizing TensorFlow Lite for Edge AI chipsets.
About this role:
We are looking for innovative problemsolvers with a passion for developing solutions that support our clients business needs. As a Machine Learning Engineer your role will be instrumental in developing and optimizing models tailored for speaker verification applications contributing to our clients mission of delivering cuttingedge solutions in speaker recognition for resourceconstrained environments.
The ideal candidate would be responsible for:
- Design implement and optimize machine learning pipelines for speaker verification.
- Collaborate closely with crossfunctional distributed teams to integrate features into edge devices.
- Conduct detailed exploratory data analysis extracting actionable insights for robust use cases in target embedded hardware platforms.
- Work closely with software engineers to deploy and finetune machine learning models ensuring optimal performance on Edge AI chipsets.
- Participate in the complete development lifecycle from model design to deployment focusing on the unique requirements of our applications.
- Ability to develop and debug productionquality code
- Familiarity with continuous integration models and unit test development
- Scientific thinking ability to design experiments and training frameworks and to outline and evaluate intrinsic and extrinsic metrics for model performance aligned with business goals.
Requirements
- Advanced degree (MSc or PhD) in Computer Science Machine Learning Speech Science or a related Data Science field.
- Proven experience developing machine learning models for voice biometrics speech modelling Speaker
- Recognition Verification Authentication Diarisation or Identification. Expertise in Python Frameworks and toolkits (TensorFlow PyTorch TFlite ONNX NumPy ScikitLearn Pandas)
- Strong understanding of signal processing techniques acoustic modeling and resourceefficient model design for speaker verification.
- Familiarity with deep learning architectures (TDNN RCNN) optimized for speaker verification in Edge AI environments.
- Excellent problemsolving skills with the ability to adapt models to meet computational constraints specific to speaker verification applications.
Skills & Qualifications: Advanced degree (MSc or PhD) in Computer Science, Machine Learning, Speech Science, or a related Data Science field. Proven experience developing machine learning models for voice biometrics, speech modelling, Speaker Recognition, Verification, Authentication, Diarisation or Identification. Expertise in Python Frameworks and toolkits (TensorFlow, PyTorch, TFlite, ONNX, NumPy, Scikit-Learn, Pandas) Strong understanding of signal processing techniques, acoustic modeling, and resource-efficient model design for speaker verification. Familiarity with deep learning architectures (TDNN, RCNN) optimized for speaker verification in Edge AI environments. Excellent problem-solving skills with the ability to adapt models to meet computational constraints specific to speaker verification applications.