Data Scientist

2d2 days ago

MyShell

Remote · Full-time

About this role

MyShell is revolutionizing the AI landscape by building an open ecosystem for AI-native apps. The company seeks a Data Scientist to design and lead a unified data and experimentation framework that resolves legacy inconsistencies across tables, metrics, and pipelines.

Day-to-day work centers on refactoring data structures, establishing company-wide metric definitions, and owning A/B testing methodology from traffic allocation through analysis workflows. The role requires close partnership with engineering on data warehouse modeling and ETL design.

The team comprises talent from MIT, Princeton, and Oxford working in a supportive environment backed by leading VCs. Collaboration spans engineering and business units to reduce decision-making costs and accelerate iteration across all product lines.

This position offers the opportunity to build standardized experiment frameworks and data governance practices that directly shape product optimization and growth strategies in a rapidly scaling AI platform.

Requirements

Bachelor’s degree or above in Computer Science, Statistics, Mathematics, Information Engineering, or related fields
3+ years of experience as a Data Scientist, Data Product Manager, Data Engineer, or Growth Analyst
Strong engineering skills with solid experimentation methodology and deep business understanding

Responsibilities

Review and refactor existing data table structures, field semantics, and key/ID systems to resolve legacy issues such as one field with multiple meanings
Design and drive a unified data model and metric definitions; establish a company-wide data dictionary and data standards
Partner closely with engineering to participate in data warehouse modeling and data pipeline design, improving data quality and maintainability
Own the overall methodology and implementation path for A/B testing and other online experiments across the company
Design experiment pipelines including traffic allocation, tracking strategy, data collection, storage, and analysis workflows
Develop standardized experiment analysis frameworks and reusable templates including core metrics, significance testing, and sample size estimation
Deeply engage with core business lines to define problems and key metrics starting from product and business goals
Promote data naming conventions, field definitions, and tracking standards while establishing data quality monitoring mechanisms