DatabricksNick Ragonese, Wesley PasfieldExplainer
Baseball teams can use Databricks to turn Statcast and team-derived pitch-level data into game-day plans for hitters, bullpen changes, pinch-hit choices, advance scouting, and roster decisions. Genie, on Unity Catalog, queries governed Delta tables and reusable Unity Catalog SQL functions for tendencies by count, handedness, pitch type, location, and base-runner state. A Multi-Agent Supervisor built with Agent Bricks and deployed on Model Serving simulates matchup pockets, compares reliever arsenals with hitter profiles, and produces explanations that staff distill into printed decision cards. For advance work, AI Search indexes pitcher and hitter embeddings to find similar arms, while Lakebase Postgres keeps reports, tags, and decisions close to the source data for shared operational memory.
The setting is a baseball organization using one governed Databricks lakehouse across clubhouse preparation, bullpen planning, late-inning offense, advance scouting, and front-office work. The workflows cover Genie, Unity Catalog, Agent Bricks, Model Serving, AI Search, and Lakebase Postgres.
Analysts use Genie with Unity Catalog tables and reusable UC SQL functions to answer matchup questions. A Multi-Agent Supervisor on Model Serving simulates game situations and compares hitter and reliever profiles, while AI Search retrieves pitchers with similar embeddings. Lakebase Postgres stores reports, tags, and decisions, and curated Unity Catalog data is published into the app.
Reusable Unity Catalog SQL functions encode tendencies by count, hand, pitch type, location, and base-runner state for repeated planning and agent workflows.
A Multi-Agent Supervisor compares reliever arsenals with specific hitter pockets and turns simulated bullpen or pinch-hit scenarios into printed decision cards.
AI Search uses pitcher and hitter embeddings in Unity Catalog to identify comparable pitchers when direct head-to-head Statcast history is limited.
The agent simulates relevant situations before the game, calls Unity Catalog SQL functions, and explains the matchup in plain baseball language. Analysts turn the output into bullpen cards, game cards, or pinch-hit grids that staff can consult without logging into Databricks.
Genie analyzes pitch mix and movement from Unity Catalog tables, while Databricks AI Search finds pitchers with similar profiles using embeddings stored in Unity Catalog. UC SQL functions then calculate lineup outcomes against those comparable pitchers.
Lakebase Postgres stores scout reports, coaching tags, and front-office decisions, making app writes immediately available to authorized analysts and agents. Scheduled jobs or pipelines also publish curated Unity Catalog data into the app so its interface has current stats and features without manual CSV exports.
Summaries are generated from the original by a language model. Check the original before relying on it.
See the idea
Agent memory is the knowledge that survives a run, not the working state inside one. This explainer separates the two, then maps the write path, the read path, and the database limits that cap recall.
Keep reading
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
One useful email. Unsubscribe anytime.
