Home > San Francisco jobs > San Francisco miscellaneous jobs

Posted: Monday, January 8, 2018 4:27 PM

Want to revolutionize finance?

The Role:

Earnest's Data team is responsible for building software and deploying algorithms that impact the key decisions of the business: risk assessment, automating decisions, predicting conversion, and optimizing our portfolio of loans. We practice full stack data science. Our modelers and engineers partner and own everything from data sourcing/ETL to model creation/testing and model deployment/maintenance.

As a modeler on the team, you will pair with engineers to develop, deploy, and maintain models that have the maximum impact on the business. You will do high impact modeling and also learn state of the art engineering. You will also work closely with business stakeholders across the company. If you want to learn something new, you will at Earnest.

Data Focused: Earnest is driven by data. We collect over 100,000 different data points from our clients. Using this data we service our customers with a full understanding of their financial potential and background. The depth of data at Earnest gives our models endless opportunity to improve our existing models and create new ones. It also requires software built specifically to ensure rapid/continuous deployment, ensure model consistency, and interacts with internal and external systems.

Autonomy: Our team decides how they work, takes ownership, and uses creative approaches to solving problems. Independent of her/his position in the organizational structure every team member is empowered to offer design and architectural recommendations to help us make the best product possible. Because we build models as services it gives the team great freedom in designing the solutions, but great responsibility in making sure they are the right solutions.

Culture: We find value in code reviews, pair programming, and paying down technical debt. We seek diverse of perspective, backgrounds, and evaluate ideas based on merit. We welcome innovative solutions and cross functional efforts. By collectively pooling our talent and experience we believe that we can create the best products and a massively scalable company.

Sample Projects:

* Creating and deploying auto-decision models

* Establishing a testing framework for new MaaS

* Creating and deploying a prepayment model

* Integrating A/B testing into an existing model service

Ideal backgrounds and expertise:

* MS/Ph.D. in CS, mathematics, statistics, econometrics or another quantitative/scientific discipline

* A proven track record of 5+-6 years in all phases of the model development process:

* requirements gathering

* data ingestion

* model building and tuning

* deployment and integration of models to the rest of our production systems

* monitoring of model results and impact

* Expert in scientific or statistical programming (R, Python). The ability to build and train models from scratch is essential.

* Experience and strong programming skills for deployment of developed models to a production environment (Python, Scala)

* Expert understanding of traditional relational databases (PostgreSQL, MySQL, SQL Server) and distributed systems (Redshift, BigQuery, Spark, Apache Hadoop)

* Desire to influence the shape and voice of the team through participating in the hiring process as well as mentorship of less experienced data scientists

* A history of successful collaboration with other departments or in cross-functional settings

* Understanding of Agile/Lean/Kanban practices

* Relevant domain experience developing software for Fintech, Banking, or related Consumer Financial Services companies

Qualified applicants with criminal histories will be considered for the position in a manner consistent with the Fair Chance Ordinance.


Associated topics: data administrator, data architect, data center, data manager, data scientist, data warehousing, hbase, mongo database administrator, sybase, teradata


• Location: San Francisco

• Post ID: 93216312 sf is an interactive computer service that enables access by multiple users and should not be treated as the publisher or speaker of any information provided by another information content provider. © 2018