At SpaceXSI, a small data science team serving a fast-growing company risked becoming a bottleneck as employees brought it straightforward questions requiring expertise. To address this, the team built Data Bot, a Team Bot in Grok Bot taught how the data science team works. Data Bot can query the warehouse, conduct analyses, and return results in each person's preferred form. Unlike personal bots that depend on what their owners know, Data Bot draws on a shared set of skills the data team maintains and improves for everyone. Teams can add Data Bot to Slack channels, where it answers questions and contributes proactively to ongoing work, and as its skills improve, everyone benefits from the same growing expertise. Data Bot receives context through skills covering metrics and analysis, including core metrics such as active users and revenue and the source tables and query logic behind them, plus guidance for investigating changes by segment or region and checking for holidays or logging changes; experiments, including A/B tests, power analyses, exposure balance and sample bias, rollouts, deep dives, and readouts; charts and dashboards, including visualization best practices and turning a chart in chat into a Databricks or Hex dashboard; and data engineering, including dbt models, pipeline alerts, and analytics event logging. Data Bot also has shared memory to retain knowledge from day-to-day use.