Growth Platform ML Engineer - Remote/Hybrid

Stripe; New York (NY); Hybride; USD 180 000 - 270 000; Plein temps

Avantages offerts par ce poste

Equity

Bonus/Commissions

401(k)

Medical, dental, and vision

Wellness stipend

Résumé du poste

Stripe is seeking a machine learning engineer to build and operate production ML systems that improve how we recommend products, content, and actions to businesses. You will own work from problem definition through deployment, working with data scientists, engineers, and product partners to translate model improvements into measurable outcomes.

You will design and evaluate models, build data pipelines, and deploy scalable solutions while ensuring reliability, latency, and cost considerations are

Qualifications

• 3+ years of industry experience in ML engineering, software engineering, or applied data science, with hands-on experience building and shipping ML models in production.
• Strong programming skills in Python and experience writing maintainable, tested production code.
• Practical experience designing, training, and evaluating ML models using frameworks such as PyTorch, TensorFlow, XGBoost, or scikit-learn.
• Experience building data or feature pipelines, proficiency in SQL, and familiarity with distributed data processing tools such as Spark or PySpark.
• A strong understanding of statistics, model evaluation, and experimentation, including the ability to recognize data leakage and distinguish offline model improvements from business impact.
• Experience deploying, monitoring, and debugging production ML systems, and evaluating tradeoffs among model quality, reliability, latency, and cost.
• Ability to turn an open-ended business problem into a technical approach and collaborate effectively with engineering, data science, product, and business partners.

Responsabilités

• Design, train, evaluate, deploy, and maintain models for recommendation, ranking, and personalized action selection across Growth Platform surfaces.
• Improve contextual bandit and policy-learning approaches, including exploration, reward design, and how recommendations adapt to user context and feedback.
• Build agent-based recommendation capabilities that use business context to identify relevant products and integration options, with evaluations that test recommendation quality and usefulness.
• Develop reliable data and feature pipelines for training and inference. Improve data freshness, feature quality, and consistency between training and production.
• Build reusable tooling for model evaluation, retraining, and safe rollout so the team can test and ship improvements faster.
• Own the quality and operation of the team's ML components: write tested production code, monitor models and pipelines, investigate failures, and improve reliability, latency, and cost.
• Design and analyze online experiments with data science partners. Connect offline evaluation to product adoption and incremental impact, with guardrails for dismissals, unsubscribes, and user experience.
• Partner with product engineering to integrate models into recommendation delivery systems, and with ML infrastructure teams to use and improve Stripe's shared training, feature, and serving capabilities.
• Work with product, marketing, and sales partners to identify problems that shared ML capabilities can solve, and make practical choices about where modeling adds value.

Connaissances

Python

ML in production

SQL

Data pipelines

Statistics

Model evaluation

Spark

PyTorch

TensorFlow

scikit-learn

Outils

PyTorch

TensorFlow

Spark

XGBoost

scikit-learn

Description du poste

Stripe is seeking a machine learning engineer to build and operate production ML systems that improve how we recommend products, content, and actions to businesses. You will own work from problem definition through deployment, working with data scientists, engineers, and product partners to translate model improvements into measurable outcomes.

You will design and evaluate models, build data pipelines, and deploy scalable solutions while ensuring reliability, latency, and cost considerations are