Descripción completa del empleo
The Role
As Data Scientist / Deep Learning Engineer in the VI (Visual Intelligence) Core Team, you will contribute to a largely self-sufficient team that develops, delivers, runs, maintains and evolves innovative machine learning powered services that provide significant value to Solera’s customers.
Visual Intelligence (VI) is Solera’s market-leading automated damage detection solution driven by state-of-the-art computer vision technology. The VI Core Team is the self-sufficient team that owns the AI technology behind VI and exposes it to its clients via a RESTful API. In addition to VI the team owns and develops ML components for other Solera products such as an invoice checking solution for the property claims space.
What You’ll Do
First Month
Get familiar with our tooling and infrastructure, our portfolio and components, the data, and the team
Run some of our services end-to-end
Make your first PR
Start contributing your ideas on all levels (data, models, code, team, etc.)
First few Months
Work on our DS components, reproducing results, understanding challenges, contributing code, and planning improvements towards our team's objectives
Synchronously and asynchronously communicate your ideas to your peers, build consensus and document progress, discussions, and decisions
Conduct experiments towards a desired outcome using a hypothesis-driven approach
Collaborate with the team to make decisions around practices and tooling and to drive the continuous improvement of the team's ability to deliver outcomes
After Onboarding
Take full ownership of one or more data science components that the team owns
Innovate through research and experimentation with new tools, methods and technology
Propose radical ideas around existing products as well as new product ideas
Dedicate part of your work time to making progress towards your personal learning objectives
What You’ll Bring
Degree in a quantitative discipline
Deep understanding of and proven ability to apply data science and machine learning concepts such as deep learning, CNN (in particular computer vision), instance segmentation, boosting, statistical analysis, predictive modelling, time series and temporal data
Demonstrated proficiency in Python and the Python data science ecosystem (SciPy, NumPy, pandas, seaborn, Matplotlib, Dask, …). Additional proficiency in second language would be a plus
Experience with data analysis and evaluation (in particular image data), to create curated and efficient datasets
Experience with image augmentation and preprocessing techniques (using PIL, OpenCV and albumentations)
Proficiency either PyTorch or TensorFlow, OpenCV, and SQL
Familiarity with any of the following would be a plus or desirable:
Google Cloud Platform (particularly Big Query, Vertex AI, Dataflow)
Kubernetes, Apache Beam
ML Operations (CD4ML)
3D graphics and synthetic training data
Concepts such as photogrammetry, vision transformers, structure from motion
NLP
Soft Skills
Proficiency in the English language (spoken and written) to be able to be successful with a globally distributed team and stakeholders
Demonstrated capacity to clearly and concisely communicate complex technical and scientific ideas and concepts
Demonstrated ability to take full ownership of your work and deliver outcomes inline with given objectives
Demonstrated ability to collaborate, hold yourself and others accountable, contribute to the advancement of the team on multiple levels and outside of primary responsibilities if necessary
Strong tendency towards communication, transparency and inclusion
Strong drive to shape the culture and practices of the team and commitment to continuous improvement of the team.