Mitek is a NASDAQ-listed global leader in mobile capture and digital identity verification solutions built on the latest advancements in AI and machine learning. Our Mobile Verify and Mobile Deposit products power and protect millions of identity evaluations and mobile deposits every day, around the world.
We are growing our Check Deposits R&D (Research & Development) Team in Spain, and we are looking for a Sr. ML Engineer like you!
The Check Deposits R&D Team is responsible for: ensuring that Check Deposits technology delivers accurate image quality assessment, check layout analysis, OCR (Optical Character Recognition) extraction of printed and handwritten characters, and check fraud detection, and last but not least, for the development of the new technology generation that will ensure the continuity of our leadership in the market by bringing check processing and fraud detection to the next level.
Your mission will be to manufacture the fastest, most accurate deposit recognition and detection engine.
As part of this team, you will have the opportunity to work with advanced computer vision algorithms and deep learning models that provide state-of-the-art performance on check processing and authentication and develop innovative technology to detect fraud on transactions based on checks.
What we offer you?
- Flexible hours and the possibility to work fully remote (in Spain) or from our Barcelona office.
- An innovative, fast-grown pace environment with a great culture.
- A diverse, inclusive, Agile-native ecosystem where everybody counts.
- Access to continuous training, participation in international conferences, joining our research reading group, and protected time for you to learn and experiment alone or in collaboration with other researchers.
- Get a compensation package that includes a highly competitive salary + bonus and many perks and benefits.
What will you do?
- As an expert in machine learning productization with a background in computer vision applied to document analysis, you will be performing the following tasks:
- Configure and tune the parameters of computer vision algorithms in the pipeline to maximize the extraction accuracy of checks, which are typically handwritten.
- Implement and optimize extraction post-processing methods that maximize accuracy by aggregating redundant extracted data leveraging confidence scores.
- Support semi-automatic data annotation for high-volume datasets by leveraging existing extraction.
- Configure and maintain feature detection algorithms on checks for fraud identification.
- Maintain and develop few-shot anomaly detection models to identify fraud on checks.
- Apply and maintain (re-training) deep learning models for field location and OCR extraction on checks.
What will you need? (to fit in this role)
- Ph.D./MSc degree in Computer Science, Electrical Engineering (or related field).
- 5+ years of experience in computer vision, machine learning, and image processing.
- Hands-on experience releasing Deep Learning models to production.
- Excellent Python and C++ programming skills.
What will be great to have to be a perfect match?
- Experience in deep learning model optimization for mobile devices and CPU inference.
- Experience with Document Image Analysis and OCR.
- Background knowledge of anomaly detection methods and statistical modeling.
- Expert on one or more machine learning frameworks: TensorFlow, PyTorch, DLib, etc.
- Experience with version control systems such as Git.
- Familiar with Agile Development methodologies.