SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 130,000 - 170,000 / annual
Taskrabbit is a marketplace platform connecting people with Taskers for home services like furniture assembly, handyman work, and moving help. You'll join a small, cross-functional discovery-stage pod focused on client retention and personalization, reporting through Product with matrix support from Data Engineering leadership.
In this hands-on individual-contributor role, you'll own the design and build of data pipelines and models that power a predictive personalization engine. Your work will directly support the team's core bet: predicting what home service a client needs and when, then reaching them proactively.
Key responsibilities include:
- Building and maintaining ELT data pipelines using modern tools (dbt, Airflow, Fivetran) that capture client home profiles, job history, and event-driven signals (weather, life events, moves)
- Partnering with the Machine Learning Engineer and Solutions Architect to prepare data models for predictions
- Developing pipelines that connect personalization signals into CRM and marketing channels (email, SMS, push, onsite)
- Creating dbt models and semantic layers to enable rapid test setup and measurement
- Leveraging AI coding tools (GitHub Copilot, Cursor, Claude Code) as a default part of your workflow to scaffold pipelines, write tests, and accelerate code review
- Contributing to technical documentation and roadmaps for scaling the data foundation
- Working daily with Product, Design, Marketing, BizOps, and ML partners in a fast-moving pod
You'll need solid experience with modern data tools (dbt, Airflow, Snowflake/BigQuery/Redshift), strong SQL and Python/Java/Scala skills, and dimensional data modeling expertise. Comfort with streaming platforms (Kafka, Kinesis) is a strong plus given the event-driven nature of predictions. Most importantly, you should regularly use AI coding assistants and know how to prompt, review, and validate AI-generated code—this isn't optional but core to how the team moves quickly from hypothesis to live test.
This role is ideal for someone energized by ambiguity, seeking outsized ownership on a small team, and wanting to help prove or kill a major product bet with real evidence. The position requires 2 days in-office at the San Francisco hub (130 Sutter St) every Tuesday and Wednesday. Note: visa sponsorship is not available; candidates must be legally authorized to work in the US.
