brewAI
Analytics help that speaks your workflow
brewAI lives inside brewlytics. It knows the functionals, reads the models, and helps you go from question to working workflow — faster than paging through documentation.
An agent for every job
Chat with your workspace
Ask brewAI about any model and it answers with receipts — reading the model, checking its last run, exploring sub-models, and mapping inputs and outputs before it replies. It can schedule tasks and find functionals too.
Auto-write model descriptions
One click documents a model: brewAI reads the graph, analyzes sub-models, and composes a description you can accept, edit in place, or send back for revision.
Generate Python
Describe what a script functional should do and the Python agent proposes a reviewable code change — apply it, or ask it to explain the code, find and fix bugs, refactor, and add docstrings.
Generate SQL
Plain English in, SQL out — scoped to your own tables. The SQL agent writes, explains, fixes, and optimizes queries right where you edit them.
AI you can measure
Usage by user and group
Track tokens, calls, and sessions for every analyst and team — filterable by agent, model, and time range.
Quality and latency
Average and p95 latency, prompt share, agent split, and model mix keep answers fast, grounded, and worth their cost.
A full audit trail
Every session is recorded — prompts, tool calls, and proposed changes — reviewable and exportable, per user and per group.