Staff Machine Learning Engineer - Retention

Taskrabbit•San Francisco, CA

3d•$170,000 - $225,000•Hybrid

About The Position

Machine Learning is a cornerstone at Taskrabbit, and we're looking for a Staff Machine Learning Engineer to join our team and lead the next phase of our customer retention strategy. This is a critical, full-stack role for an individual who is passionate about the end-to-end lifecycle: from initial research and model development to building the robust systems that power repeat customer engagement and lifetime value growth at scale. Taskrabbit's greatest growth opportunity lies in deepening customer relationships and accelerating repeat purchases. Our most valuable customers are those who return frequently, discover new service categories, and increase their spending over time. There's significant untapped potential in the marketplace: repeat customers spend 3-5x more than one-time users, and category expansion unlocks new revenue streams within our existing customer base. This role is central to capturing that opportunity. While initial matching quality and service discovery matter, the real competitive advantage and growth lever is optimizing the experience after a successful first job.

Requirements

BS, MS, or PhD in Computer Science, Statistics, Operations Research, or a related quantitative field.
8+ years of industry experience building and deploying high-quality, production-grade machine learning models and systems.
Strong theoretical knowledge and hands-on experience in machine learning, particularly in search, ranking, recommender systems, pricing/elasticity modeling, or predictive analytics.
Solid software engineering skills with proficiency in one or more programming languages, including Python.
Experience with popular ML libraries like Scikit-learn, lightgbm, xgboost, TensorFlow, PyTorch, etc.
Proficiency in SQL is also required for writing complex queries and transforming data.
Experience building REST API-based services.
Experience with modern data and ML technologies, such as Docker, Kubernetes, Kafka, Airflow, data warehouses (eg snowflake, redshift or BigQuery), and data lakes.
Familiarity with tools for Infrastructure as Code, such as Github actions, and CI/CD pipelines.
Excellent communication skills, with the ability to present complex findings and recommendations clearly to both technical and non-technical audiences.
A passion for quickly learning new technologies and a drive to solve challenging problems, and a collaborative mindset.

Nice To Haves

Familiarity with dbt is a plus for transforming and testing data.
Ideally, experience working in marketplace or platform contexts where ranking, matching, and pricing directly impact user experience and business outcomes.

Responsibilities

Own the reliability and performance of our core ranking system, ensuring accurate tasker-to-job matching and optimizing First-Time Right (FTR) rates.
Increase repeat purchase frequency through intelligent matching, personalized recommendations, and category discovery.
Expand customer lifetime value by helping customers find and return for new service categories.
Optimize affordability and relevance via dynamic pricing, smart segmentation, and category-specific experiences.
Reduce friction and churn through predictive quality interventions and proactive customer success.
Build marketplace resilience by developing systems that keep high-value customers engaged and loyal.
Own the complete lifecycle of models—from feature engineering and training through evaluation, deployment, monitoring, and optimization in production.
Build and maintain scalable, reliable ML infrastructure and data pipelines that support reproducible feature engineering and model deployment across real-time, near real-time, and batch contexts.
Develop monitoring and observability systems to understand data quality and model performance in complex systems.
Collaborate with engineering and science teams to optimize algorithms for training, inference, and evaluation.
Write clean, efficient, and maintainable code.
Participate actively in code reviews, documentation, and best practices across the full software engineering lifecycle.