At Ocrolus, we believe companies work best when they focus on their core business and let automation do the rest. We’re powering the digital lending ecosystem and help financial services firms make high-quality decisions with trusted data and unparalleled efficiency.
Ocrolus’ Human-in-the-Loop document automation software analyzes documents with over 99% accuracy. We're replacing legacy OCR vendors that cap out at 75-80% accuracy, and augmenting the robotic work that humans are prone to doing all too often – which can be expensive, error-prone, and slow. By empowering lenders to analyze diverse sources of financial data more efficiently, Ocrolus levels the playing field for every borrower, providing expanded access to credit at a lower cost.
We’ve raised over $100 million from blue-chip investors and are working with customers like PayPal, Brex, SoFi, Blend and Plaid . Join us as we build the future of fintech, and make an impact at an award-winning, high-growth startup that Forbes recently dubbed the “ Next Billion-Dollar Startup ”.
Summary
The data science team at Ocrolus builds high-quality, impactful machine-learning based products that empower lenders to make better credit, fraud, and operational risk decisions. Data scientists play a critical role in the full product development cycle and move fast to ideate, build, deploy, and maintain production quality models. If you are a data scientist looking to leverage your strong engineering abilities in building ML models end-to-end, then we want to talk to you!
What you’ll do
- Collaborate with product and engineering to identify client needs/market opportunities, translate them into data problems, and build scalable SaaS solutions
- Engineer features across multiple sources of structured and unstructured data
- Independently develop stable and efficient ML models that balance complexity with interpretability/product deadlines
- Deploy, monitor, and maintain production quality models that are queried thousands of times each day by our clients
- Examples of data science initiatives include: predicting loan default probability and loss-given-default based on transactional data, using NLP to classify pay stub fields into standardized categories, and building an entity resolution system to match financial documents across time with a specific borrower
What you’ll bring
- 5+ years working as a Data Scientist or MS with 3+ years of experience in a quantitative discipline (e.g., math, statistics, computer science, engineering)
- Full stack data-scientist experience: ideating, building, deploying, monitoring, and maintaining production ML models that solve product needs and perform with high levels of accuracy, stability, and coverage
- The ability to communicate and present complex technical topics and results to various audiences
- Passion for understanding the “why” of the problem and the impact of solutions on client outcomes
- Deep understanding of statistics, probability, and machine learning algorithms
- Strong software engineering and data engineering fundamentals
- Production quality programming skills in Python
- Excellent SQL skills and comfort working with large and complex data warehouses
- Experience with CI/CD, shell scripting, Git/version control, and APIs
Bonus points
- Background in financial services (e.g. business lending, consumer lending, credit cards)
- Portfolio of past data science accomplishments (including source code)
We take pride in our dynamic, diverse team, unified by shared values of Ownership, Optimism, Objectivity, Humility, Urgency, and Appreciation . We love what we do and the people we do it with, which is why we welcome every individual, provide them with equal opportunity irrespective of their race, gender, gender identity, age, disability, national origin or any other legally protected rights that one has.
We look forward to hearing from you!