Soteris is a YC-backed, seed-stage, profitable company building a machine-learning-based insurance pricing platform. Earlier this year, we finished a 15-month pilot with an auto insurer that resulted in almost tripling their policy profitability. As a result, in less than a year, our revenue has tripled. Additionally, in the last month we’ve signed a number of new customers up for pilots, which has the potential to quadruple our revenue further from here. The opportunity per customer is so large because our ML tech integrates directly into the critical path of their business (pricing the risk).
There’s $750B of insurance policies written in the US every year, yet insurance companies generally lack the sophistication necessary to price insurance individually. Instead, they generally place people into a small number of discrete buckets that dictates the price charged. Our ML platform allows them to calculate a highly customized price instead. The team is under 5 people, remote across the US, and growing. You would be our second ML engineer and would be responsible for building out both the ML infrastructure we use to train insurance-pricing algorithms from application + claims data, as well as the models themselves.
WHAT YOU'LL DO:
As an ML Engineer at Soteris, you will get to work on many facets of the ML functionality that drives business value for our customers. Expect to dive right into several of the following:
• Onboarding new customers’ into our model-training process.
• Improving and extending our ML models to predict policy cash flow over time.
• Enhancing our automated model monitoring to watch for things such as performance degradation and data drift.
• Pursuing opportunities for further automation of our training / evaluation pipelines by enabling customers to seamlessly integrate into our systems.
• Creating and automating additional data analytics and reports we offer our customers on top of their application and claims data.
• Researching additional lines of insurance to help drive our expansion roadmap.
Here's a look at our current tech stack. Experience with these is ideal, but it’s not required so long as you have the ability and desire to pick them up quickly:
• Fully cloud-based on AWS
• SageMaker for ML workflows - Data exploration, processing, model prototyping, training, management, and serving
• Airflow for job orchestration
• Amazon API gateway for exposing public APIs
• AWS Lambda Functions (Python) for general purpose utility functions
• ML and data processing code in Python (Lightgbm, Pandas, scikit-learn)
• Python for backend services code
• Infra managed by Terraform
If this seems interesting to you and you'd like to find out more, please reach out by clicking the Apply button!
Machine Learning Engineer, TenyksFull Time Employment
Lead Machine Learning Engineer, HackerRankFull Time Employment
Machine Learning Engineer, Sapling.aiFull Time Employment
Machine Learning Engineer (MLOps), BelvoFull Time Employment
Senior NLP Machine Learning Engineer, Topline ProFull Time Employment