Data Scientist/Machine Learning Engineer

Traditional HR software gives teams better tools. We're different: our AI/ML systems do real, strategic work — designing compensation structures, running merit cycles, creating offer letters.

What you'll do:

• Choose and apply the right statistical, optimization, and machine learning techniques to build proprietary models

• Partner with domain experts to understand their empirical knowledge, then provide scientific grounding and validation for what already works in practice

• Balance autonomy with collaboration: you'll have the freedom to test new approaches, but also the humility to absorb knowledge from CompExecs, engineers, and the broader team to validate ideas and drive the best business outcomes

• Drive data-driven storytelling and decision-making to shape product direction and strategic choices

• Independently evaluate model performance, design data pipelines, and ensure we have the right infrastructure for analytics at scale

• Champion a data-informed culture by establishing best practices and shaping the strategic role of data science at Comp

The outcome: Data-driven insights that help Comp deliver expert-level strategic HR work at the speed and scale of software.

• 3+ years of experience in quantitative roles (data scientist, ML engineer) operating in highly ambiguous environments, ideally at hyper-growth product companies
• Strong proficiency in SQL and Python for data analysis and statistical/ML/optimization modeling
• Strong communication skills, with the ability to partner effectively with product managers, engineers, executives, and domain experts
• Strategic mindset capable of generating insights that go beyond statistical significance to inform long-term product and business direction
• Thrives in unstructured environments with a high degree of autonomy

Nice to have:

• Experience at early-stage or hyper-growth startups
• Experience with or knowledge of AI/LLMs, particularly evaluating generative AI systems
• Background in optimization or advanced statistical methods