Machine Learning Researcher

Rise Talent Advisors; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Benefits offered by this job

Equity

Job summary

Rise Talent Advisors is helping a venture-backed company in San Francisco hire a machine learning researcher to work on-site five days a week. You will build predictive models for underwriting and financing decisions using traditional ML and deep learning on large, real-world datasets that impact real financing decisions.

The role is about 70 percent research and 30 percent production. You will design the research plan, perform rigorous evaluations with temporal validation and leakage checks,

Qualifications

• 5 to 12 years of machine learning research experience.
• Master’s or PhD in computer science, math, physics, or a quantitative field.
• Experience building predictive models with traditional ML and deep learning.
• Rigorous evaluation habits and production deployment experience.

Responsibilities

• Build predictive models for underwriting and financing using ML and DL.
• Work with large real-world datasets and enforce strict evaluation (temporal validation, calibration, leakage checks).
• Perform feature engineering and interpretability work with SQL, Pandas/Polars, SHAP.
• Design the research plan: questions, holdout, metrics, success criteria.
• Put models into production and monitor drift after launch.
• Split time between research and production to ensure robust models.
• Explain model behavior and limits to partners who use results for decisions.

Skills

ML Research

Deep Learning

Predictive Modeling

Model Evaluation

Production ML

Data Analysis

Education

MSc/PhD in CS/Math/Physics or quantitative field

Tools

SQL

Pandas/Polars

SHAP

Job description

This role is on site in San Francisco, five days a week.

We are helping a venture backed company hire a machine learning researcher for its San Francisco office. The company uses machine learning to support financing decisions for consumers and businesses. You would build the predictive models behind its underwriting and financing systems, working with large, messy, real-world datasets that affect actual financing decisions, not only offline experiments.

The work is about 70 percent research and 30 percent production. This is classical machine learning and deep learning for tabular and structured prediction, not large language model wrapper work. The company is hiring for one seat.

What you will do

• Build predictive models for underwriting and financing, using both traditional machine learning and deep learning.
• Work with large, messy, real-world datasets and keep evaluation rigorous, including temporal validation, calibration, and checks for leakage.
• Do feature engineering and interpretability work with tools such as SQL, Pandas or Polars, and SHAP.
• Design the research plan before a model is trained: the question, the holdout, the metrics, and what would count as a real improvement.
• Get models into production, then watch how they behave after launch and retrain when the data drifts.
• Split your time between research and production so the models that ship are ones you would stand behind.
• Explain model behavior and limits clearly to partners who use the results to make financing decisions.

What you need

• 5 to 12 years of machine learning research. Academic lab work counts.
• A Master's or PhD in computer science, math, physics, or a quantitative field, preferably from a top 30 university.
• Experience as an ML researcher, research scientist, or data scientist doing machine learning research and development.
• Experience building predictive models with traditional machine learning and deep learning.
• Rigorous evaluation habits, plus feature engineering or interpretability work.
• Fast career progression.
• You can work on site in San Francisco five days a week.
• Visa transfers such as OPT and H-1B can be considered.

Nice to have

• Experience modeling messy real-world systems, such as in robotics, physics, or chemistry.
• A first-principles approach to hard problems.
• Credit or lending knowledge is not required.

What this search is not

• This search is not for SQL-only data science profiles.
• This search is not for people whose work is mainly large language model wrappers or prompt-only work.

Location

• On site in San Francisco, California, five days a week.
• Full time, direct hire role. Equity is part of the offer.

Rise Talent Advisors is an equal opportunity employer. We welcome people of all backgrounds.