Traackr - Data Scientist

• Partner with Product and Engineering to identify high-impact opportunities, frame ambiguous problems, define success metrics, and choose pragmatic approaches (heuristics, statistics, ML, or GenAI). • Lead rigorous experimentation across teams: hypothesis design, metric/guardrail definition, power analysis, A/B testing (or quasi-experiments), and clear readouts that drive decisions. • Build and iterate on ML/AI capabilities that ship to production (e.g., classification, information extraction, ranking/recommendations, and GenAI components such as RAG or developing the agent harnesses for our core agentic journeys), optimizing for value-added, latency, and cost. • Establish best-in-class evaluation practices for both ML and LLM features: golden datasets, offline/online evaluation plans, regression suites, and monitoring that catches quality drift early. • Enable engineers to build safely and effectively with AI by coaching on prompt patterns, tool/function calling, structured outputs, guardrails, and debugging/evaluation workflows. • Design and support agentic workflows where they add real product value, with clear constraints, observability, and fallbacks. • Support the end-to-end lifecycle of deployed models and AI systems: data requirements, training/fine-tuning where relevant, validation, deployment, monitoring, incident response, and continuous improvement. • Raise org-wide leverage by creating reusable assets (evaluation harnesses, shared datasets, templates, documentation) and running enablement workshops. • Communicate insights and tradeoffs clearly to technical and non-technical stakeholders, turning analyses into decisions and measurable impact. • Champion responsible, privacy-aware AI: appropriate data handling, bias/fairness considerations where applicable, and human-in-the-loop workflows when needed.