Principal Data Scientist

Role Overview

• End-to-end data infrastructure: From ingestion to orchestration to serving (batch/stream), optimizing cost, reliability, and speed.
• Data quality, integrity, and availability: CI/CD for data, contracts, tests, and observability as first-class citizens.
• Pipelines & processes: Design/optimize ingestion, transformation, and analysis workflows for structured and unstructured data.
• Insights to action: Partner with GTM, Product, Operations, and Partner Success to prioritize questions, run experiments, and land changes.
• Models & semantics: Own core metrics, semantic layers, reporting models, and feature stores aligned to the partner funnel (SQL→MQL→SQL→ARR).
• Governance & collaboration: Lineage, catalog, permissions, and clear documentation so teams can self-serve safely and confidently.
• Design/optimize pipelines (ingestion, transformation, orchestration).
• Develop and maintain analytical and ML/LLM models (structured/unstructured; stats/experimentation).
• Implement/improve dashboards (Dash/Plotly, Power BI, Tableau, Looker).
• Analyze trends/patterns to improve processes, product, and the partner funnel.
• Data modeling (dimensional, lakehouse, feature stores).
• Data quality & observability (detect/fix inconsistencies).
• Coordinate accessibility & usability (catalog, lineage, permissions).
• Automate extraction/transformations, data testing, and deployment.
• Document best practices, standards, and methodologies.

Requirements

• A principal level: 8+ years across Data Science/Analytics/Engineering (including leading initiatives), with shipped data products and measurable business impact.
• Mastery of data tooling: Python (preferred) and/or R; SQL & NoSQL; visualization (Power BI, Tableau, Looker, Dash/Plotly); ML/LLM/stats; cloud data services.
• You use AI responsibly to speed up quality work.
• English skills: Intermediate level required — Level B1 minimum (B2+ is a plus) to collaborate with international teams.
• Data SQL & NoSQL excellence: Schema design, performance tuning, partitioning, indexing.
• ETL/ELT & orchestration: Reliable pipelines (batch/stream) with monitoring, alerting, and recovery.
• Modeling: Dimensional modeling, lakehouse patterns, and feature stores that serve analytics and ML.
• ML/LLM & experimentation: From baseline models to rigorous A/B tests and causal inference on structured/unstructured data.
• Visualization: Power BI, Tableau, Looker, and Dash/Plotly for decision-ready views.
• Observability & governance: Data tests, lineage, contracts, catalogs, and CI/CD for data.
• Quality by design: Freshness, completeness, accuracy, and SLAs as non-negotiables.
• Engineering System design & architecture: Comfortable with monoliths, microservices, and event-driven patterns.
• APIs & services: Build/consume services with Flask/FastAPI; productionize models and data apps.
• Cloud & data services: Practical fluency with cloud storage/compute and managed data offerings.
• Core stack: Python, Pandas, Polars, DuckDB, scikit-learn; DL with PyTorch/TensorFlow/HuggingFace.
• DevX: Linux, bash, Git, containerization; automation, testing, and CI/CD for code and data.
• Security-minded: Design with least privilege, encryption, and secure defaults.

Tech Stack

• Cloud
• ETL
• Flask
• Linux
• Microservices
• NoSQL
• Pandas
• Python
• PyTorch
• Scikit-Learn
• SQL
• Tableau
• Tensorflow

Benefits

• Private health insurance with access to a top-tier medical network.
• A friendly, inclusive culture where everyone can be themselves.
• Casual dress code (yes, t-shirts and sneakers are perfectly fine).
• Team gatherings and informal events to celebrate milestones and recharge together.