On-site and live remote
AI coding workshop for engineering teams
Geist Labs comes to your team, audits how engineers use AI coding tools today, turns the findings into a practical adoption plan, and runs a hands-on workshop on your real repositories.
It is an agentic coding workshop for engineering teams: practices, guardrails, and operating artifacts your engineers keep — not a tool demo they forget by the next sprint.
the opportunity
Same team. Same tools. Same development cost. 2-3X the output.
Most engineering teams run their AI coding tools at a fraction of what those tools can already do. The limit is rarely the model, the budget, or the headcount—it is the workflow around them. No one has shown the team how. That is what this workshop is for.
The workshop is tailored before anyone enters the room.
You get a current-state AI coding audit, a prioritized adoption plan, and an on-site, hands-on workshop built around your team's repositories and delivery workflow.
FIG W.1
Audit
Map how engineers use AI coding tools today across repositories, delivery workflows, review, and measurement.
FIG W.2
# adoption-plan.md
└─ priorities
└─ owners
└─ measures
Plan
Turn the findings into a prioritized adoption plan with outcomes, owners, guardrails, and a tailored agenda.
FIG W.3
hands-on team workflow
Workshop
Work together on your real repositories so the team leaves with shared practices and operating artifacts.
who it is for
AI training for engineering teams, aimed at the people responsible for adoption
The strongest fit is a team already experimenting with coding assistants or agents but lacking a shared, measurable way of working.
Engineering leaders
Create a focused adoption roadmap, success measures, governance, and a rollout cadence the team can own.
Software engineers
Practice reliable agentic coding workflows on the repositories, tools, and review standards used every day.
Platform and enablement teams
Standardize context, workflow gates, evidence, and measurement across teams without forcing one coding tool.
the engagement
Audit, plan, then run the workshop
The agenda is based on evidence from your environment, not a generic training deck.
stage 01
Current-state AI coding audit
Understand how the team uses AI today before prescribing training.
We review the team's tools, selected repositories, delivery workflows, context practices, review standards, usage patterns, and adoption blockers. The audit identifies where AI is helping, where it creates risk or rework, and where a shared workflow can create the most leverage.
- Stakeholder and practitioner discovery
- Repository, workflow, and review-readiness assessment
- Opportunity and risk map ranked by impact
current state assessment
01 stakeholder + engineer discovery
02 workflow walkthroughs
03 selected repository review
findings ranked with your team
stage 02
Team-specific adoption plan
Translate the audit into decisions, owners, outcomes, and a workshop agenda.
We turn the findings into a practical plan: which workflows to improve first, what context and instructions agents need, where human review stays essential, what the team will practice, and how progress will be measured after the session.
- Prioritized workflow and capability roadmap
- Tailored agenda, exercises, and repository examples
- Success measures for quality, delivery, adoption, and cost
priority 1 context routing · owner assigned
priority 2 agent review loop · owner assigned
priority 3 delivery evidence · owner assigned
measures: quality · cycle time · cost
stage 03
On-site AI coding workshop
Your team learns the operating model by using and building it.
Geist Labs facilitates a hands-on session at your office—or live remote when preferred. The team practices planning, context routing, agent delegation, code review, verification, and handoffs on representative work, then turns those practices into artifacts it can keep using.
- Exercises grounded in your tools and repositories
- Agentic coding workflows: plan, implement, review, verify
- Repository instructions, review gates, and operating loops
hands-on session
exercise 01 plan + context routing
exercise 02 agent delegation + review
exercise 03 verification + handoff
artifacts owned by your team
engagement deliverables
What your team leaves with
Concrete decisions and artifacts your team can use after the workshop—not just notes from a training session.
Current-state audit
A clear view of current AI usage, workflow gaps, repository readiness, review risks, and adoption blockers.
Prioritized adoption plan
A sequenced plan for the highest-value workflows, with owners, guardrails, measures, and next steps.
Repository-ready artifacts
Context maps, instructions, workflow contracts, review gates, and examples built around your environment.
Shared team operating model
A practical way to plan, delegate, review, verify, measure, and improve AI-assisted engineering work.
Want to start before the engagement? These are the same artifacts the workshop installs, published for individual engineers to use today.
- AGENTS.md starter templates — The repository instructions the workshop builds on.
- Reliable agents need gates — The review gate teams practise during the session.
- Reduce AI coding costs — What the audit usually finds first.
delivery
Built around your office, stack, and engineering reality
The engagement can be delivered on-site with your engineering team or live remote. Scope and timing are set after the current-state audit so the workshop addresses the workflows and decisions that matter most.
Facilitator
Workshops are led by Brandon W. Lee, founder of Geist Labs and the lead researcher behind its AI Systems Engineering frameworks.
Format
On-site at your office or live remote
Material
Your selected repositories, workflows, tools, and standards
Focus
Reliable adoption, shared practices, quality, cost, and evidence
questions teams ask
AI coding workshop questions, answered directly
What engineering leaders usually need to know before planning the engagement.
What is an AI coding workshop for engineering teams?
It is a hands-on team engagement for adopting AI coding tools in real software delivery. Geist Labs first audits current usage, builds a tailored plan, and then trains the team on its own repositories, workflows, review standards, and goals.
Can the workshop be delivered on-site at our office?
Yes. Geist Labs delivers the AI coding workshop on-site with your engineering team. Live remote delivery is also available when that is a better fit.
What happens before the workshop?
Geist Labs audits how the team uses AI today, reviews selected repositories and delivery workflows, identifies gaps and high-leverage opportunities, and turns those findings into the workshop plan.
Do we work on our real codebase?
Yes. The workshop uses selected repositories and representative engineering workflows so the practices, instructions, review gates, and operating artifacts fit the way your team actually ships software.
Is this an agentic coding workshop for engineering teams?
Yes. The workshop can focus on agentic coding workflows such as task planning, context routing, bounded delegation, code review, verification, and evidence-based handoffs. The audit determines which practices matter most for your team.
Which AI coding tools does the workshop cover?
The engagement is tool-aware but not tied to one vendor. It can be tailored to the coding assistants and agents your team already uses while focusing on durable practices for context, planning, review, verification, and measurement.
How is this different from generic AI training for engineering teams?
Generic training teaches features or prompt techniques. This engagement starts with your current state and leaves behind a team-specific plan, repository-native artifacts, shared workflows, and clear next steps.
What does the engineering team leave with?
The team leaves with a current-state assessment, prioritized adoption roadmap, tailored workshop materials, repository-ready operating artifacts, and a practical cadence for continued improvement.
Start with a free introduction to your team, goals, and current AI coding workflow.
Not ready for a workshop but want to start implementing the system? Read the AI Systems Engineering Handbook — the same operating model the workshop installs, written to work through on your own.