Staff Machine Learning Engineer
Taskrabbit3 months ago
Base Salary
$170k - $225k/yr
Responsibilities
- Own the architecture, model design, experiment strategy, and production reliability of Taskrabbit’s core ranking system.
- Act as the primary technical decision-maker and technical interface for Product, Data Science, Risk, and Commerce on matching and ranking.
- Build automated retraining pipelines, rollback capabilities, and skew detection to support safe experimentation and prevent silent model drift.
- Improve ranking quality by addressing position bias and other systemic biases in the ranking pipeline.
- Use client intent signals, return-client signals, Tasker preferences, and preference modeling to improve conversion and personalization.
- Identify broader ML opportunities, make build-versus-buy recommendations, and contribute to long-term ML architecture.
- Mentor junior engineers, elevate senior engineers through code and design reviews, and raise the team’s technical bar.
- Introduce new methodologies, conduct technical R&D, and provide thought leadership on ML problems and future ranking capabilities.
Requirements
- BS, MS, or PhD in Computer Science, Statistics, Operations Research, or a related quantitative field.
- 8+ years of industry experience building and deploying production-grade machine learning systems.
- End-to-end ownership experience for a production system, with deep expertise in ranking, recommender systems, or two-sided marketplace ML.
- Experience with debiasing, large-scale A/B experimentation, feedback loops, data pipelines, feature stores, training infrastructure, serving, and monitoring.
- Strong Python and SQL proficiency.
- Hands-on experience with PyTorch, TensorFlow, scikit-learn, LightGBM, or XGBoost.
- Experience with Kafka, Airflow, Snowflake or BigQuery, Docker, and Kubernetes.
- Experience with Infrastructure as Code using Terraform and CI/CD pipelines.
- Familiarity with dbt is preferred.
- Ability to conduct critical R&D, make architectural decisions, communicate technical tradeoffs, and mentor engineers.
- Proficiency using AI coding tools such as Claude Code, Augment, or similar tools across the software development lifecycle.
Benefits
- Hybrid schedule requiring two days per week in the San Francisco office on Tuesdays and Wednesdays.
- Employer-paid health insurance for US employees.
- 401k match with immediate vesting for US employees.
- Flexible paid time off and two company-wide closure weeks.
- Product, wellness, productivity, and education stipends.
- IKEA discounts and reproductive health support.
- Benefits vary by country of employment.
- Visa sponsorship is unavailable; candidates must already be authorized to work in the United States without employer sponsorship.
Tech Stack
Apache AirflowApache KafkadbtDockerGoogle BigQueryKubernetesLightGBMPythonPyTorchscikit-learnSnowflakeSQLTensorFlowTerraformXGBoost
Categories
About Taskrabbit
Taskrabbit builds a two-sided marketplace and mobile app that connects consumers with local Taskers for home services such as furniture assembly, moving help, and handyman work. The company generates revenue through service fees on bookings and partnerships, and was founded in 2008 and is headquartered in San Francisco. Taskrabbit operates across North America and Europe and has been a subsidiary of IKEA since its 2017 acquisition.