Deep dive
From your first prompt to your first agent.
Lucy meets you where you are and coaches you up a ladder, on your real work, with depth in the skills your role actually uses.
In short
Lucy teaches ten universal capability areas, then specialized depth for your role, up to solution architecture and agent design. She keeps the coaching current by tracking AI tools daily, and she teaches you to match the shape of the solution to the shape of the problem.
The foundation
Ten universal capability areas.
The capability ladder
Climb at your own pace.
Not everyone needs to reach the top. Most people get a lot from the first two rungs, and small gains across a whole team add up. Your plan starts from your goal, even before Lucy has observed anything, and works back to the next step you can actually take. Plus domain depth: a developer goes deep on code generation, a finance analyst on data analysis, a sales rep on pipeline work. Nobody burns time on techniques they will never use.
Foundation
Setting up your AI tools safely, writing good prompts, giving AI the right context, checking what comes back.
Tool fluency
Finding the approved, specialized AI tools that fit your kind of work, and using them well. Lucy tracks the new ones so you don’t have to.
Solution architecture
Framing the problem, combining tools, designing agents, building in rules and checks, reusing what works. You don’t need to be technical to climb it.
Solution shapes
Match the shape of the solution to the shape of the problem.
One of the most important skills Lucy builds in you. A task is one-off (prepare for this meeting); a solution keeps working (always be prepared for meetings). Builders think in solutions.
From lightest to heaviest:
A method you learn once: designing an agent
Role · goal · tools · sources · output · stop condition. Learning it once is what makes your second build faster than your first.
Always current
You never have to keep up with AI news again.
Behind Lucy, a living map connects roles, goals, tasks, workflows, tools and skills, and the most useful AI tools are tracked daily, so what Lucy coaches today works today.
Tool categories Lucy coaches across
Lucy coaches generative AI tools with natural-language interfaces, inside your approved tool list. That boundary keeps the coaching deep.
Where it leads
After about six months, a Solution Builder (someone who builds AI solutions for their own work) can…
Spot which parts of your work to hand to AI, and which to leave alone
Pick the right solution shape for each problem and build it
Compose approved tools into governed, reviewable workflows
Redesign the process around AI instead of pasting AI onto it
Turn something you built once into a solution your team adopts
Questions
Questions about what Lucy teaches
Which AI tools does Lucy coach?
Lucy coaches generative AI tools with natural-language interfaces, inside your approved tool list. That covers chat assistants, embedded copilots, code copilots, meeting assistants, workflow automation, data and BI tools, agent frameworks and more. The most useful AI tools are tracked daily, so what Lucy coaches today works today, and keeping to your approved list keeps the coaching deep.
What will I learn first?
Whatever your work needs first. Most people start with the foundation, applied to a real task from their Personal Coaching Plan: prompting, iteration, checking output, managing context, choosing the right tool and safe use. Then comes tool fluency and, at your own pace, solution architecture: framing problems, combining tools and designing agents. Depth follows your role, so a finance analyst goes deep on data analysis and a developer on code generation.
How long does it take to become a Solution Builder?
Think in months of practice on real work, not a weekend course. After about six months, a Solution Builder can spot which parts of their work to hand to AI, pick the right solution shape and build it, compose approved tools into governed workflows, redesign a process around AI, and turn something they built into a solution their team adopts. Everyone climbs at their own pace.
Will Lucy teach me to build an AI agent?
Yes, when the problem calls for one. An agent fits work that repeats and takes several steps, like assembling a brief before every meeting. Lucy teaches a method you learn once: role, goal, tools, sources, output and stop condition, with clear limits on what the agent may do. That’s what makes your second build faster than your first. Often, though, a prompt or a workflow is the better answer.
What happens if my company switches AI tools?
Lucy’s coaching moves to your new approved tools, and your skills carry over. Breaking down tasks, managing context, checking output and choosing the right solution shape work across tools and model generations, so you apply the same skills in the new tool instead of starting over.
Keep exploring
More deep dives
Ask Lucy what she’d teach you first.
Lucy will happily explain herself, and then ask about your work.
What’s here today vs. what’s coming
Some features described on this site, like proactive coaching, are still being built.
