Data Scientist

Surge; United States; À distance; USD 150 000 - 230 000; Plein temps

Avantages offerts par ce poste

Bonus opportunity

Health, dental, and vision insurance

401(k) plan with employer match

Unlimited PTO

Résumé du poste

Surge is seeking a senior individual contributor to embed within product development. You will transform proprietary mortgage data into scores, recommendations, and patterns used in customers’ daily workflows.

You will work with engineering to ship features that appear in the product, using AI daily, while validating outputs with human-in-the-loop governance and transparent decision documentation.

Qualifications

• Graduate training ideal in economics, econometrics, statistics, data science, or related field.
• Evidence of ability ranks higher than degree; credential gatekeeping is not required.
• Comfort with data prep, modeling, validation, and shipping to production.

Responsabilités

• Build statistical, econometric, and ML models for customer-facing capabilities.
• Design scores, indicators, and decision aids with audit-ready rigor.
• Move between exploratory analysis and production deployment.
• Collaborate with engineering and analytics teams on what to measure and surface.

Connaissances

Data science

Modeling

Validation

Shipping

Communication

Formation

PhD or equivalent in quantitative field

Outils

Snowflake

Python data stack

Claude Code

Salesforce

AWS S3

Google Workspace

Description du poste

Surge is an AI-first software and data company serving the wholesale mortgage lending space. We run Partner360, a partner relationship management platform with a ten-year production history and a six-year zero-outage record, built on proprietary mortgage-industry data that nobody else can assemble. We recently entered a new chapter under new ownership, and we are building the team that takes us through our next phase of growth.

This is a senior individual contributor role embedded inside product development. Not a research seat. Not a dashboard seat. Your work ships to paying customers and shows up in revenue. You will work alongside engineering to take proprietary mortgage and customer data and turn it into the scores, recommendations, and surfaced patterns that customers see inside their day-to-day workflow.

What You Will Do

Model and Method

• Build the statistical, econometric, and machine-learning models that drive customer-facing capabilities
• Design quality scores, risk indicators, and decision aids that hold up to audit, customer scrutiny, and capital-markets due diligence
• Run causal-inference and pricing-theory work where the question demands it, not just predictive modeling
• Move fluently between exploratory analysis and production deployment

Product-Embedded Work

• Pair with engineering and customer-facing teams on what to measure and what to surface
• Translate concrete customer needs into a model, a score, or a pattern-detection script
• Ship features that show up in the product, not in a report
• Co-develop metrics with sophisticated customer-side analytics counterparts when the engagement calls for it

AI-Forward Execution

• Use AI tools daily for code, analysis, documentation, and review
• Build pattern-detection work that runs in the background and pushes insights, not chatbots that wait to be asked
• Verify AI-generated output. Human-in-the-loop is the gate
• Apply AI to real product problems, not just exploration

Communication Hygiene

• Document decisions, assumptions, and uncertainty: what we know, what we estimated, what we have not yet measured
• Work async-first in a distributed team environment
• Surface blockers immediately rather than hoping they resolve
• Build artifacts a colleague can pick up cold

What We Are Looking For

• Graduate training ideal (PhD, PhD-track, or equivalent applied experience) in economics, econometrics, statistics, data science, computational social science, or a closely related quantitative field
• We also welcome experienced professionals with equivalent backgrounds. Evidence of ability ranks higher than the degree, and we do not gatekeep on the credential
• Comfortable with the full stack of applied work: data preparation, modeling, validation, and shipping
• A daily AI tool user, not "AI-curious"
• Comfortable owning a model end-to-end, from question to production
• Statistical foundations strong enough to argue about identification strategy, not just AUC
• Can explain a model to a non-technical executive in language they recognize
• Mortgage, fintech, or capital-markets background a strong plus, not required

The job lives most heavily in Snowflake (data warehouse and modeling surface), the Python data stack, and Claude Code (daily AI-assisted analysis). It also touches Salesforce data, AWS / S3, and Google Workspace. Snowflake fluency, or equivalent modern cloud data warehouse experience and a willingness to pick it up fast, is the main technical anchor.

Who You Are

• An individual contributor who wants to build products people use, not papers that get cited
• An async communicator who documents decisions and context
• Transparent about what is measured versus what is estimated versus what is still open
• Willing to be wrong publicly and update fast

How We Hire

As part of the interview process, you will spend a working session demonstrating how you actually work with AI tools on a real applied-data problem, live, in a notebook, against data we hand you. No pseudocode, no trivia. We are evaluating problem framing, model selection, validation discipline, and how fluently you use AI as a collaborator.

Compensation and Benefits

• Competitive compensation based on experience (including bonus, 401(k), etc.)
• Health, dental, and vision insurance
• Unlimited PTO

Work Location

Remote, United States.

All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, age, disability, sexual orientation, national origin, or any other category protected by law.