Senior Machine Learning Engineer (Voice Biometrics)

Hi! Discover the possibilities of cooperating with Scalo!

At Scalo, we focus on delivering software projects and supporting our partners in developing their business. We create software that enables people to make changes, act faster, and achieve better results. We utilize a wide spectrum of IT services to support our clients in achieving their business goals.

Our areas of activity include, among others:

technological consultancy, software development, embedded systems, cloud solutions, data management, dedicated project teams.

What do we offer as part of our cooperation?

the opportunity to provide services in diverse projects in the areas of Software, Embedded, Data, and Cloud Services,

involvement in additional pre-sales and expert initiatives,

access to medical care, a sports card, and a cafeteria platform in accordance with the applicable cooperation rules.

Daily tasks

• implementation of a project regarding the development of modern voice biometrics solutions used for user authentication and fraud prevention in a contact center. The system utilizes Machine Learning, Voice Biometrics, and AWS for real-time risk analysis and supports organizations in more effectively protecting customers from fraud,
• technology stack used in the project: Python, AWS, Amazon Kinesis, AWS Fargate, Amazon SageMaker, TensorFlow, PyTorch, Scikit-learn, SQL, NumPy, Pandas, Docker, GitHub, CI/CD,
• designing, developing, and optimizing solutions for speaker verification and speaker identification,
• developing processes for embedding extraction, enrolment, scoring, calibration, and threshold optimisation,
• analyzing audio quality and optimizing models for noise, codecs, and various devices,
• implementing anti-spoofing and liveness detection solutions,
• deploying and maintaining production Machine Learning systems in an AWS environment,
• building and developing data pipelines using Amazon Kinesis,
• deploying models using AWS Fargate and SageMaker Endpoints,
• developing monitoring, observability, and optimizing system performance,
• participating in code review, pair programming, and sharing knowledge with the team,
• 100% remote model,
• a rate of 170 PLN/h on B2B depending on experience.

Requirements

Competencies required to perform the services:

min. 4-5 years of commercial experience in Machine Learning projects related to voice biometrics,

practical experience with speaker verification and speaker identification solutions,

knowledge of issues related to embeddings, scoring, model calibration, enrolment, and anti-spoofing,

experience with Python and TensorFlow and/or PyTorch frameworks,

practical knowledge of AWS services, in particular Kinesis and Fargate and/or SageMaker,

knowledge of MLOps, monitoring, deployment, and maintenance of production models,

good knowledge of SQL, Git, Docker, and CI/CD environments,

ability to work with existing systems and develop production solutions,

ability to cooperate effectively in a team and participate in code review and knowledge sharing,

English language proficiency at a minimum of C1 level.

Must have: Machine Learning, Python, TensorFlow, PyTorch, AWS, Kinesis, AWS Fargate, MLOps, SQL, Git, Docker