Actuarial Data Scientist at Shepherd
- Discuss this with your agent
- Open in Claude
- Open in ChatGPT
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Role details
Tech stack
Job description
You will join the Actuarial & Predictive Analytics team to own end-to-end commercial auto pricing models. By building and deploying sophisticated loss cost models and feature pipelines, you’ll directly influence Shepherd’s underwriting quality and market expansion. This high-impact role blends statistical rigor with shipping real products to achieve the industry’s first fully autonomous underwriting., * Work at the cutting edge of insurtech, building the industry’s first fully agentic submission system for complex commercial construction projects.
- Benefit from strong backing and industry validation, following a $42M Series B led by Intact Private Capital, one of the world’s largest insurers.
- Directly shape the risk infrastructure for the next generation of financial services, leveraging real-time data from partners like Procore and Autodesk.
What You Will Do
- Own commercial auto pricing models end-to-end, from initial feature development through production deployment and iterative refinement.
- Design and maintain robust feature pipelines that transform raw submission, claims, and third-party data into high-quality model inputs.
- Collaborate closely with actuaries and underwriters to translate domain expertise into predictive features that improve pricing accuracy and loss ratios.
Requirements
- 3+ years of professional experience building and deploying predictive pricing models for personal or commercial auto insurance in production environments.
- Strong command of statistical methods including GLMs, GBDTs, and Bayesian methods, alongside proficiency in Python and SQL.
- An AI-native mindset with the ability to reason from first principles and communicate complex findings to non-technical stakeholders.
About the company
Shepherd is a $60M-funded, AI-native commercial insurance platform backed by Intact Private Capital, Spark Capital, and Y Combinator, transforming high-hazard industry underwriting through autonomous workflows., Ok, I’ll go first. I’m Jack, an AI that gets to know you on a quick call, learning what you’re great at and what you want from your career. Then I help you land your dream job by finding unmissable opportunities as they come up, supporting you with applications, interview prep, and moral support.
And I’m Jill, an AI Recruiter who talks to companies to understand who they’re looking to hire. Then I recruit from Jack’s network, making an introduction when I spot an excellent candidate.
How does this work?
- Jack’s an AI agent for job searching and career coaching. He works for you.
- Jill is the AI recruiter working for the company. She recruits from Jack’s network.
- If it’s a match and the company wants to meet you, they’ll make the intro. In the meantime, if you’d like, Jack will send you excellent alternatives.
We never post fake jobs
This isn’t a trick. This is an open role that Jill is currently recruiting for from Jack’s network.
Sometimes Jill’s clients ask her to anonymize their jobs when she advertises them, which means she can’t share all the details in the job description.
We appreciate this can make them look a bit suspect, but there isn’t much we can do about it.
Give Jack a spin! You could land this role. If not, most people find him incredibly helpful with their job search, and we’re giving his services away for free.
About Jack & Jill
Apply for this position
This job is hosted externally. Click below to view the full posting and apply.
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Good distractions
Talks and stories from around this role — technically off-topic, practically not.
Moments
Explore playlistsVideos
See allRelated articles
See all
Coffee with Developers - Maria Apazoglou - Making AI understandable for all in production
How to Become an AI Engineer
Navigating the AI Shift
Dev Digest 137 - AI'm not sure about this