Senior Growth Data Scientist

Role overview Build the evaluation and reliability program for an AI copilot embedded in a multi-asset trading platform. The work spans benchmark design, release safeguards, model improvement, and operational monitoring, with a focus on making market analysis, portfolio reasoning, and trading actions accurate and safe for experienced investors.

Responsibilities - Create evaluation systems for correctness, safety, response time, and regression risk across analysis, risk reasoning, and trading workflows. - Maintain benchmark datasets, scenario suites, stress tests, and adversarial cases that reflect changing market conditions. - Establish automated quality checks that prevent releases when important measures deteriorate. - Work with product and engineering teams on predictable, auditable action flows, including previews and confirmations. - Improve models through curated data, labeling and quality review, error analysis, and changes to prompts or tools; use model training approaches when they can improve measured results. - Set up AI telemetry, alerts, incident investigation, and processes for resolving problems quickly. - Build fluency in trading concepts such as margin, short selling, portfolio risk, and execution so the copilot can explain them clearly.

Requirements - At least eight years delivering production software and strong programming ability in at least one language. - Solid computer science foundations, testing practices, and systems design skills. - Experience building evaluation frameworks, test harnesses, and benchmarks for complex systems such as language models, agents, search, or ranking systems. - Experience managing model improvement cycles, from dataset curation and labeling through offline experiments and measured deployment. - Ability to define metrics and measurement pipelines, then use results to guide product and engineering decisions. - Comfortable debugging across model and service tooling, adding instrumentation, and collaborating on user experience patterns that support safety and trust. - Self-directed and willing to learn unfamiliar areas to solve problems.

Nice to have - Fine-tuning, preference optimization, distillation, or other methods for improving tool-use reliability. - Domain-specific or adversarial benchmarks for high-stakes systems. - Knowledge of trading across asset classes and margin types; Rust or performance-sensitive services experience. - Experience setting incident response practices and service objectives for AI systems.

Benefits and work setup Compensation includes a stated base salary range of $180,000–$250,000, with potential additional compensation such as bonuses or equity. Benefits include equity, retirement plan matching, parental leave, and medical, dental, and vision coverage. The listing also mentions in-office meals and kitchen amenities and identifies the role as remote.