Data Scientist

About Semgrep

Semgrep, the leader in code security for builders, empowers invention without friction. Teams catch, flag, and fix real issues before they ship, powered by security that learns as they build. Semgrep secures code as it’s written and provides guardrails that pave the road for developers to move fast and stay secure. Built for builders and trusted by security, Semgrep lives where developers work, delivering fixes without breaking flow, and giving security teams visibility, control, and confidence.

Semgrep gets smarter as you build, with AI that learns your context to cut false positives and prioritize reachable vulnerabilities, validated by 95% of security reviewers across 6M+ findings. Semgrep makes zero false positives a reality with App Sec teams triaging 80% fewer false positives across Code and Supply Chain, dramatically shrinking the backlog.

Founded in San Francisco and backed by Menlo Ventures, Felicis Ventures, Lightspeed Venture Partners, Redpoint Ventures, and Sequoia Capital, Semgrep is recognized by Gartner in Application Security Testing and is trusted by leading organizations, including Snowflake, Dropbox, and Figma. Learn more at semgrep.dev
.

You will be an early member of Semgrep’s data team. Your mission will be to define how an entire company uses data, always striving to best improve our users’ security. You will work on a diverse set of problems, touching every aspect of the startup: extracting product insights from usage metrics, determining business strategy from market data, crafting production data pipelines, and defining where to direct our security research.

This is a growth role: while you will start as an individual contributor, initially contributing to a quarter-long project, you will grow your technical skill set and domain knowledge to start taking on more responsibility and influencing data and business decisions within the company.

Along the way, you will work with a dedicated group of full-stack, backend, and infrastructure engineers, as well as security researchers and program-analysis developers. You will learn what it means to have “secure-by-default” code, meet and collaborate with security-industry scions, and be part of the decisions that make a high-growth startup successful. Your work will be critical to our mission and every feature you build will have a measurable impact on our users’ lives.

• Contribute to specific data science projects and initiatives at Semgrep, discovering each department’s most pressing data problems, and proactively identifying the most critical areas to focus your efforts.
• Build dashboards to track board-level metrics, apply multivariate regression to identify important product features, and use active-learning techniques to guide data collection and labeling.
• Iteratively tackle problems as a series of experiments, proving the value of your work with proof-of-concept to ever more refined results.
• Convince your peers of your conclusions with clear data visualizations and well-reasoned explanations.
• Help grow your team through the recruitment and hiring of top data talent.

• 2+ years of experience in data and strategy fields.
• Knowledge of data-science approaches, including machine-learning algorithms, optimization methods, symbolic AI, statistical methods, and the taste to know when to use each.
• Experience clearly visualizing information and experimental results across the full company stack: board-level, leadership team, and individual team leads.
• Familiarity with production data-processing pipelines and tools such as S3, Five Tran, DBT, Snowflake, Metabase, Retool, Sagemaker/Jupyter Notebook (Python).
• Aptitude delivering technical projects via rapid iterative development.
• Experience working on a small team in a fast-paced environment and willingness to experiment with different approaches before settling on the best solution given time constraints.
• Excellent, proactive communication, both verbal and written.

• Build a client-facing dashboard showing scan time metrics over time.
• Work with Product leadership to identify the…