1 /

Your AI Doesn't
Know You

How to build an AI system that actually knows your business

Ashkaan Hassan · We Solve Problems · EO Los Angeles

The Status Quo

You open ChatGPT.

You explain your business from scratch.

You get a decent answer.

You close the tab.

Repeat tomorrow.

Sound familiar?

Why It Feels Underwhelming

AI without context is like hiring a brilliant consultant who has amnesia every morning.

You're doing all the work to bring them up to speed.

Every. Single. Time.

The AI is smart — the problem is that it doesn't know you.

The Cost of Staying at Level 1

You repeat yourself constantly

Every conversation starts from zero. Your business model, your clients, your standards — explained again and again.

Output is generic

Without your voice, your brand, your preferences — everything comes out sounding like it was written by a stranger. Because it was.

Nothing compounds

You can't ask "what did we decide last month?" because there is no last month. Every session is isolated.

You're the bottleneck

Nothing happens unless you open the chat. No scheduled tasks, no monitoring, no proactive work. The AI waits for you.

These aren't AI limitations. They're setup limitations. And they're fixable.

The AI Maturity Spectrum

L1
Chatbots
Ask a question, get an answer, start over.
L2
Persistent Context
AI knows your business, your people, your standards.
L3
Automated Workflows
AI runs on a schedule, monitors systems, takes action.

Most people are at Level 1. This talk is about getting to Level 2 and 3.

The Key Insight

The latest AI model with no context about your business will always lose to an older model that knows everything about you.

Scaffolding beats raw intelligence.

So how do you give AI your context?

Part I

Building Your
AI's Brain

The six pieces of an AI operating system

Piece 1

Instruction Files

Your AI behaves the same way every time — your way.

Think of it as an employee handbook. It covers how you communicate, what tools to use, what rules to follow, and what's off limits.

You write it once. Every conversation after that follows your rules.

# My AI Instructions
Be concise, not verbose
Technical depth is welcome
Never send emails without my approval
Use sub-agents for research
End every session with a journal entry
Commit and push all changes

What Goes in an Instruction File

Your Voice

Output matches your tone — direct, concise, technically deep. No more editing AI drafts to sound like you.

Your Team

The AI delegates research, bulk edits, and scheduled tasks to specialist agents. It manages them like you'd manage a team.

Your Workflow

Every session follows the same lifecycle: load context, do the work, log what happened, save everything. Automatically.

Your Safety Net

The AI never sends emails without approval, never exposes credentials, never takes an action you haven't authorized.

Different AI agents can get different instruction files — a research agent gets different rules than your primary agent.

Pro tip

The Context Router

The AI always has exactly the right information for what you're doing — without you asking for it.

Writing an email? It already loaded your email templates. Talking about a client? It already pulled their profile.

It never loads everything at once. It stays fast and focused by only pulling what the moment requires.

# When this happens → load this
Writing emails → email templates
Building apps → dev style guide
Project work → project index
Person mentioned → their profile
Using an API → API docs
Handling credentials → security rules
# Only loads what's needed, when needed
Piece 2

Your Rules & Preferences

Everything comes out in your voice — on the first draft, every time.

Your email style, code conventions, presentation format, communication tone — captured once, applied forever.

No more editing AI output to sound like you. It already does.

Email Templates
How to structure proposals, follow-ups, cold outreach
Code Conventions
Naming, structure, testing standards
LinkedIn Voice
Tone, topics, formatting for posts
Behavioral Rules
Guardrails, safety policies, deploy procedures
Piece 3

Your Domain Knowledge

The AI knows your world — your people, your numbers, your goals. It never asks "what do you do?"

👥
People
Team, clients, contacts, family
📈
Business
KPIs, processes, strategy
💰
Finance
Budget, tax, investments
🎯
Goals
Quarterly targets, milestones

Mention a person — the AI knows them. Discuss finances — it has your budget. Context shows up automatically.

Piece 4

Your Automations

Work happens while you sleep. Reports generate, systems get monitored, content gets drafted — all on a schedule.

The AI understands each automation and can modify, debug, and extend them on its own.

I have dozens of automations running right now — from news digests to financial reports to health tracking.

# Example automations
Morning news digest (daily 7 AM)
Weekly business scorecard (Mon midnight)
Security assessments (per client)
Goal tracking dashboard (every 15 min)
LinkedIn content publisher
System health monitoring (24/7)
Piece 5

Your Integrations

Your AI talks to all your tools — pulling data, pushing updates, and orchestrating across systems you already use.

Google Workspace
QuickBooks
1Password
Cloudflare
Todoist
Home Assistant
Autotask
NinjaOne
Hudu
Strety
Monarch
LinkedIn
TRMNL
Zoom
Notion

Your AI becomes the nervous system tying all your tools together.

Piece 6

Long-Term Memory

"What did we decide about the pricing model?" Six months later, the AI knows.

Every initiative is tracked as a project with status and next steps. Every session ends with a journal entry logging decisions and changes.

The AI's memory gets better over time — not worse.

Projects

Each initiative gets a folder with status, description, and next steps. The AI updates these as work progresses.

Journal

Daily session logs capture every decision, change, and lesson. Searchable forever. Every session compounds on the last.

The Complete Picture

Instruction Files

Consistent behavior, every time

Rules & Style

Everything in your voice

Domain Knowledge

It knows your world

Automations

Work happens while you sleep

Integrations

All your tools, connected

Long-Term Memory

It remembers everything

All plain text. All in one folder. All version-controlled.

No database. No proprietary format. No vendor lock-in.

Part II

Your AI Team

One orchestrator, multiple specialists

The Delegation Chain

Your primary AI doesn't do everything itself. It delegates to specialists — just like you manage a team.

Primary Agent (Claude)
Plans, coordinates, writes, journals — the "manager"
Research Agent
Web research, video summaries, fact-checking
Code Agent
Bulk edits, file changes, code generation
Workflow Engine
Scheduled jobs, monitoring, alerts

Each specialist has its own instruction file. The primary agent only receives a summary back — keeping it fast.

A Typical AI Session

01
You start a conversation — AI loads your preferences and rules automatically
02
You describe the task — AI loads only the context it needs (router)
03
AI does the work — delegates to specialists, calls your integrations
04
You review and approve — human in the loop for anything that matters
05
AI journals the session — logs decisions, changes, and what's next
06
Everything is saved — version-controlled, searchable, permanent

Every session builds on every session before it. That's the compounding effect.

Part III

“Show Me the Money

What this system produces in practice

Weekly Scorecard — Zero Data Entry

BEFORE

Every Monday, 45 minutes pulling ticket data, calculating KPIs, typing numbers into a spreadsheet.

AFTER

Ticket system → AI computes KPIs → pushes to scorecard. Every Monday at midnight. Zero human effort.

0
minutes per week
was 45 min

Morning Briefing

Every day at 7:00 AM, an email lands in my inbox:

Curated news digest (filtered for relevance)
Today's calendar at a glance
Task priorities for the day
Weather and commute info
7:00
Every. Single. Day.

I didn't ask for it. It just shows up.

Self-Healing Automation

Dozens of automations running 24/7. When something fails:

01

A monitoring system catches it immediately

02

AI diagnoses the root cause

03

Applies the fix automatically

04

Notifies me only if it can't self-resolve

Most failures are fixed before I even notice.

The Real ROI

Weekly scorecard
45 min/week0 min
Security assessments
hours/clientminutes
Morning briefing
15 min/day0 min
Content production
hours/piecereview only

This isn't incremental improvement. This is a step change.

And it compounds — every automation you build frees time to build the next one.

Part IV

Your Turn

How to start building this for yourself

Start This Week

1
Write your instruction file (30 minutes)

Tell the AI who you are, how you communicate, and what rules to follow. Save it where your AI can read it.

2
Add your business context

Start small: your team, your key clients, your quarterly goals. Even a few files make a massive difference.

3
Have a contextual conversation

Ask your AI something about your business. Watch the difference when it actually knows you.

Where Are You Today?

L1

“Hey ChatGPT, help me write an email.”

L2

AI knows your business. Every conversation builds on the last.

L3

AI runs on a schedule. Takes action. Maintains itself.

Getting to Level 2 takes an afternoon. Level 3 is the project after that.

Want a Head Start?

I open-sourced the framework behind everything in this talk. It's called Contextium.

Pre-built folder structure
Integration connectors for popular tools
Sample automations to learn from
Works with Claude, Gemini, Cursor, Copilot

Apache 2.0. Free forever. You own everything.

terminal
$ curl -sSL contextium.ai/install | bash

One command. Five minutes. You're at Level 2.

Q & A

Open floor

Thank You

Ashkaan Hassan

We Solve Problems · wesolve.tech · EO Los Angeles

contextium.ai — open source, free forever