A Kavanah Course — Free · 12 modules · Beginner to expert

Mastering Kavanah

Sixty-eight lessons that take you from your first login to running a workspace an AI agent can operate — every page, every setting, and the workflows that connect them.

The product manual, taught as a course. It walks the whole application surface by surface: projects and the task board, clients and the deal room, chat, mail and calendar, time, planning, reports, the knowledge base, and the KVN charter. Then it goes deep on the AI agent — what it can actually do, how to talk to it, how to govern it with autonomy levels and action policies, and how to hire AI Employees that run recurring work on their own. It ends with integrations, the developer surfaces, workspace administration, and the daily and weekly rhythms an expert operator actually runs.

68 lessons ~12 hr 39 min total68 of 68 published

An Introduction

How to Learn Kavanah

What this course covers, how it is ordered, and how to read it with the app open.

Listen

0:00 / ~4:10

Kavanah is a large application. It runs projects and tasks, a CRM and a sales pipeline, internal chat, email, a calendar, time tracking, planning, reporting, a knowledge base — and an AI agent that can touch all of it on your behalf. That breadth is the point: the agent is only useful because the work already lives in one place. But it does mean the product cannot be learned by clicking around for an afternoon, and nobody should have to. This course is the manual, ordered so that each thing you learn makes the next thing make sense.

What this course is, and what it is not

This is the nuts and bolts. Where each page is, what is on it, what every control does, what to type, what to click, and what happens after you do. Where a feature has a trap — a setting that is off by default, a default that surprises people, an action that cannot be undone — the lesson says so plainly rather than leaving you to find it in production.

It is not a management course. Kavanah ships one of those too — Project Management with Kavanah, in the same Learn section — and the two are deliberately separate. That course argues about why estimation works the way it does, what to measure, and how to delegate to a system that acts. This one shows you which button produces the estimate. Where a mechanic here only makes sense against the theory, the lesson links across rather than re-arguing it.

How it is organized

Twelve modules, sixty-eight lessons, in the order a real user meets the product.

Modules 1 to 2 are day one: signing up, the shape of the app, moving around it quickly, and the projects and tasks that everything else hangs off.

Modules 3 to 6 are the rest of the surface area: people and permissions, clients and deals, the communication surfaces, time and planning and reporting, and the knowledge and charter tools that give the agent something to reason from.

Modules 7 and 8 are the AI agent — two modules, not one, because using it and governing it are different skills. Module 7 is what it is, how to talk to it, what it can genuinely do, and how to read back what it did. Module 8 is the machinery you use to let it act: autonomy levels, action policies and approvals, AI Employees, runbooks and goals, scheduled runs, exec meetings, and the analytics that tell you what it is costing.

Modules 9 to 11 are the connective tissue and the admin: integrations and how to keep them healthy, the build surfaces (automations, the remote browser, apps, the API), and running the workspace — plans and credits, security, directory sync, trash and audit.

Module 12 is the expert rhythm: what a day and a week actually look like when the product is set up well, how to design a workspace an agent can run, how to roll it out to a team, and what to do when something breaks.

How to read it

With the app open in another tab. Every lesson ends with a Kavanah-in-practice block whose steps link straight to the surface being described, and with a short list of what to watch to know the thing is working. Doing the steps as you read is worth more than reading twice.

Read modules 1 and 2 in order — later lessons assume you know what a task and a project are in Kavanah's specific sense. After that, the modules are mostly independent: jump to the one covering the surface you are stuck on. The two agent modules are the exception and should be read together, in order, because module 8 configures what module 7 describes.

If you are an admin setting up a new workspace, read 1.6 (the first-week checklist) early — it is the short version of everything, with pointers to the lesson that explains each step properly.

The fastest way to be good at Kavanah is not to memorize where things are. It is to understand which surface owns which decision, so you always know where to go and what the agent will find when it looks. That is what this course is for.

Module 1

Getting Started

Signing up, finding your way around, and the setup that pays for itself

6 lessons · ~1 hr 7 min

Your first hour and your first week. The shape of the app, how to move around it without the mouse, what your profile actually drives, how to get your team in, and the setup checklist that decides whether everything after this works.

Module 2

Projects and Tasks

The board, the task, and the queues that feed them

7 lessons · ~1 hr 21 min

The core of the product. Creating and structuring projects, the three board views, every field on a task and what it drives, subtasks and dependencies and recurring work, custom statuses and properties, your personal queue, and the triage queue where captured work gets accepted or dismissed.

Module 3

People, Clients and Deals

Who is in the workspace, what they can see, and who you are selling to

5 lessons · ~59 min

The people directory and the skills that drive assignment, roles and per-project access, the client record and the client portal, the deal room with its pipelines and cadences, and two-way Google Sheet sync for deals and accounts.

Module 4

Communication

Chat, mail, calendar, intake, and the notifications that tie them together

6 lessons · ~1 hr 1 min

Every surface where words arrive. Internal chat with threads and handles, the unified email inbox across Gmail, Outlook and ProtonMail, the calendar, public request pages for intake from outside, workspace announcements, and how to tune notifications so they stay readable.

Module 5

Time, Planning and Insight

Where the hours go, what fits in the sprint, and what the numbers say

6 lessons · ~1 hr 6 min

Time tracking and the timer, how estimates are produced and calibrated, the planning surface and its horizons, the cross-project portfolio, the report library with scheduled delivery, and the recurring State of the Business email.

Module 6

Knowledge and Thinking Tools

The documents, the canvases, and the charter that governs the agent

4 lessons · ~43 min

The knowledge base with its decisions, glossary and playbooks, whiteboards and mind maps for thinking in shapes, and the KVN charter — the one piece of writing that changes what the AI agent will and will not do.

Module 7

The AI Agent: Using It

What it is, how to talk to it, and how to check its work

5 lessons · ~58 min

The agent as a user of your product rather than a chatbot bolted onto it. What it can genuinely reach, how to phrase a request so it lands, how to dictate instead of type, how to read the ledger of what it did, and how its memory and context actually work.

Module 8

The AI Agent: Governing It

Autonomy, approvals, AI Employees, and recurring autonomous work

7 lessons · ~1 hr 22 min

The machinery that makes an acting agent safe to run. The four autonomy levels and the two-sided ceiling, action policies and the approval queue, hiring AI Employees with scoped capabilities, runbooks and goals and the waiting-on ledger, scheduled runs, exec meetings, and the analytics that show what it all costs.

Module 9

Connecting Kavanah

Bringing your other tools in, and keeping the connections alive

6 lessons · ~1 hr 8 min

How the integration model works and what connecting actually grants, the Google Workspace suite, Microsoft and Slack, the project tools, GitHub and the Dev surface, and the maintenance nobody plans for — expiring grants, broken-connection alerts, and how to tell a real breakage from a false alarm.

Module 10

Build and Extend

Automations, the cloud browser, apps, and the API

5 lessons · ~54 min

The surfaces for making Kavanah do something it does not do out of the box: no-code automation rules, the shared cloud browser with saved logins, the app marketplace, publishing your own app, and driving the whole product from the REST API or the command line.

Module 11

Running the Workspace

Settings, money, security, provisioning, and recovery

6 lessons · ~1 hr 5 min

The admin surface, end to end. A map of every settings tab, how per-seat plans and credits actually bill, account security and two-factor, SCIM provisioning and domain auto-join, trash and undo and the audit log, and Manager Mode for holding yourself to your own commitments.

Module 12

Expert Kavanah

The rhythms, the workspace design, and the troubleshooting

5 lessons · ~55 min

What it looks like when it is all working. The daily and weekly loops an expert operator runs, how to design a workspace the agent can actually operate, how to roll the product out to a team without losing them in week two, and a diagnostic order of operations for when something silently does not happen.