Pick the right tasks
Teams learn how to break work into agent-sized loops: investigate, plan, patch, test, review, and land.
Maude Agent Enablement
One sprint to make coding agents count
Maude Agent Enablement is a focused sprint for teams already experimenting with coding agents. In one sprint, developers learn how to plan work, load context, steer agents, review patches, and use session history to improve the next loop.
One sprint, one shared way to ship with agents
Turn scattered usage into team habits your developers can search, share, and reuse.
Practice
Task framing, context loading, prompting, tool use, and stopping rules.
Reviews
Live review of prompts, plans, diffs, tests, and outcomes from your repos.
Playbook
Team norms for PRs, eval tasks, prompt search, escalation, and token-efficient delivery.
The sprint turns agent use into team execution
The gains come from repeatable habits: choosing the right work, giving the agent the right context, reviewing patches well, and learning from previous sessions.
Review generated code with confidence
Developers learn practical review routines for generated patches: diffs, tests, invariants, security, and repo conventions.
Spend tokens on progress
Teams learn how to reduce context bloat, repeated retries, vague prompts, and long-running loops that fail to move the change forward.
A curriculum built around one real sprint
The sprint uses your codebase, tools, review process, session history, and day-to-day development flow. Developers practice on real work, and platform teams leave with a clearer operating model for agent-assisted delivery.
Day 01
Choose the work and set the baseline
Pick representative tasks, review current agent habits, and identify where time, tokens, review effort, and routing choices are getting wasted.
Output: a one-page waste audit
One sprint is all it takes to supercharge your team
One-sprint program
Agent Enablement Sprint
A guided sprint for one engineering cohort: live practice, repo-specific workflow review, coding-agent office hours, session-memory setup, and a team playbook.
- Baseline: identify where agents help, stall, route poorly, or waste tokens in your current workflow.
- Practice: hands-on loops for planning, prompting, patching, testing, and review.
- Review: live critique of prompts, plans, diffs, tests, and outcomes from your repos.
- Playbook: team conventions for context, prompt search, PRs, evals, escalation, and token discipline.
Make the next sprint your agent sprint
Maude gives the enterprise a gateway it can control. Agent Enablement helps developers turn that control and session history into faster delivery and lower cost per accepted change.