Saved $24,000 and built an AI-powered organization in one year

WALKING THE TALK AWARD - PROJECT DESCRIPTION

AI as My Thought Partner: Building an AI-Powered Organization

The Transformation

A year ago, I knew very little about AI. Today, I use Claude, ChatGPT, and Claude Code as my daily thought partners—not for occasional tasks, but for strategic thinking, problem-solving, and building systems that run our entire operation.

As Head of Organizational Development at Fifth Avenue Properties—a Canadian homebuilder scaling from 117 to 919 annual home starts this year. In order to get to our BHAG of 5000 units per year I recognized that I have to be able to scale effectively and to do so I have embedded AI into every layer of how we work. This isn't experimentation. This is how we operate.

Achievement #1: The Data Foundation

The Problem: Our company data lived in silos—Power BI dashboards nobody used, spreadsheets tracking the same information differently, manual data entry across disconnected systems. You couldn't get a straight answer without hours of digging or going to several people.

The Solution: I consolidated everything into Airtable and architected it as our central operational database. I connected Claude and Claude Code directly via MCP, giving my AI thought partner real-time access to query and analyze our live data.

What I Built:

- Main Data Job Tracker: Projects, customers, contracts, deposits, margin analysis across all communities

- Preconstruction Performance Dashboard: Tracking days from firm to permit (currently 157 avg), permit processing times (40 days), jobs in preconstruction (17)

- Executive KPI + Rocks Tracker: 31 leading KPIs with status breakdowns and an AI Insights Dashboard

- Airtable forms for sales and construction staff to enter data directly—eliminating manual entry errors and creating real-time cross-team visibility

The Impact: We're saving at least $2,000/month through faster data retrieval alone. Questions that used to take hours to answer now take seconds. Sales can see construction progress. Construction can see sales status. Everyone works from the same truth.

Tools: Airtable, Claude (via MCP), Claude Code, Chat GPT, VS Code

Methodology: Single source of truth architecture. Relational database design with junction tables. Role-based views so each team sees what they need. Forms for data entry to eliminate errors.

Achievement #2: The $24,000/Year Integration Win

The Problem: Our construction management software (BuildPro) vendor wanted $2,000/month for an API connector to Airtable.

The Solution: Working with Claude as my thought partner, I designed a workaround:

1. Configured BuildPro to email automated reports

2. Built an n8n workflow to intercept and parse the Excel attachments

3. Automated data extraction and transformation into Airtable

4. Created analysis dashboards from the synchronized data

The Impact: $24,000/year saved—with more flexibility than the vendor solution would have provided. Futher, we have acutal insight into what the build times are as well as the point points. I can use Claude to analyze the data from airtable in real time. Saves hours of analyzing spreadsheets.

Tools: n8n, Airtable, Claude

Methodology: Build vs. Buy decision framework. When a vendor wants $24K/year, ask: "Can we build a creative workaround?" Sometimes the answer is yes.

Learning: This taught me that vendor solutions aren't always necessary. AI as a thought partner helps you see alternatives you'd miss on your own.

Achievement #3: Meeting Intelligence Automation

The Problem: We have four standing meetings every week. Notes scattered everywhere. Action items lost. No institutional memory.

The Solution: I built an automated pipeline that captures everything with zero manual effort:

1. Meeting occurs → Read AI automatically captures transcript

2. n8n Workflow #1: Read AI webhook → Extract data → Create Notion page → Append full content

3. n8n Workflow #2: Schedule trigger → Search Notion database → AI summarizer (using Claude API) → Route to Airtable by meeting type

4. AI extracts action items and creates tasks automatically

Next Phase: Connecting via MCP to Asana so tasks extracted from meetings automatically appear in team members' inboxes. No more manual task creation from meeting notes.

The Impact:

- Hours saved weekly—no manual meeting processing required

- Searchable company knowledge base growing automatically

- Team members more effective—tasks appear with full context

- Complete institutional memory captured and accessible

Tools: Read AI, n8n, Notion, Airtable, Claude API

Methodology: Trigger → Process → Store → Analyze → Act. Design automation pipelines that handle the boring stuff so humans can focus on thinking and decisions.

Achievement #4: Two Second Brains

The Insight: Personal learning and company knowledge need different homes. Mixing them creates noise.

What I Built:

For Fifth Avenue (Notion):

- Company Second Brain storing meeting archives, SOPs, team documentation

- Knowledge base that grows automatically from meeting automation

- Searchable institutional memory accessible to the team

For Me (Obsidian):

- Personal Second Brain following Tiago Forte's PARA methodology

- Processing all 46 lessons from AI Mastery 2025 into searchable knowledge base

- Map of Content (MOC) structure for organized retrieval

- Smart Connections AI plugin for semantic search across notes

- Syncing across PC and iPhone for mobile capture

Tools: Notion (company), Obsidian (personal), Smart Connections plugin, Wispr Flow (voice-to-text capture)

Methodology: Building a Second Brain framework. PARA folder structure (Projects, Areas, Resources, Archive). Separate containers for separate purposes.

Learning: Two Second Brains are better than one. Keeping personal synthesis separate from company knowledge keeps both clean and purposeful.

Achievement #5: Creative AI Applications

Training Videos: I created AI-powered training videos using Synthesia for employee onboarding—with a company theme song I generated using Suno.

Application Development: Using Lovable for AI-powered application development.

Visual Content: MidJourney for marketing visuals and presentation graphics.

Tools: Synthesia, Suno, MidJourney, Lovable

Achievement #6: Team Training & Multiplication

The Philosophy: AI as a thought partner shouldn't be limited to me.

What I Do:

- Weekly AI training sessions for entire staff

- Teaching practical application of Claude, ChatGPT, and other AI tools

- Hands-on training: team members build in Airtable, use AI as their thought partners

- Progressive skill building across sales, construction, and administrative roles

The Result: I've trained a team member to work in Airtable who now uses AI as their thought partner. This multiplies my impact—I'm not the bottleneck anymore.

Methodology: Teach practical application, not theory. Have people build real things. Create AI literacy across the organization.

The Thought Partner Difference

The most important thing I learned from AI Mastery 2025 is this: AI isn't a tool. It's a thought partner.

I don't just ask Claude to do tasks. I think with Claude:

- Strategic problem-solving and analysis

- Organizational design (recently: what does it take to scale from current state to 5,000 doors/year?)

- System architecture and database design

- Document creation (SOPs, job descriptions, investor materials)

- Research and competitive analysis

Right now, I'm dictating this application using Wispr Flow. AI is woven into every aspect of how I work.

What's Next

This Quarter:

- CRM-Airtable API integration (completing data unification)

- Asana MCP integration (automatic task distribution from meetings)

- Customer survey forms sent automatically, with AI-powered analysis of responses

Future:

- AI income streams targeting $50K ARR

- AI-powered customer communication service for other builders

- Construction industry AI training based on everything I've learned

- Create a Selections app for homeowners to select items in their new house

Complete Tool Stack

AI Thought Partners: Claude, Claude Code, ChatGPT Automation: n8n, Airtable + MCP Meeting Intelligence: Read AI Knowledge Management: Notion (company Second Brain), Obsidian (personal Second Brain), Smart Connections Content Creation: MidJourney, Suno, Synthesia Development: Lovable Productivity: Wispr Flow

Key Learnings

1. AI as thought partner changes everything — Using AI for collaborative thinking, not just tasks, is the unlock.

2. Start with data consolidation — You can't use AI effectively if your data is scattered. Getting everything into Airtable was the foundation.

3. Automate the boring stuff — Meeting notes, data entry, task creation. Free humans for thinking.

4. Build don't buy (sometimes) — The $24K/year BuildPro savings taught me vendor solutions aren't always necessary.

5. Train others early — Having team members who use AI as thought partners multiplies your impact.

6. Two Second Brains — Separating personal learning from company knowledge keeps both purposeful.

7. Forms are underrated — Getting staff to enter data via forms eliminated errors and created real-time visibility.

8. Speed of retrieval matters — All data in one place means answers in seconds, not hours.

9. Connected System - Created a system that is connected using API

Relevant URLs

www.fifthavehomes.com, fifthaveproperties.com

Closing

Having known very little about AI before I started this course, I now have a solid roadmap for creating value in my company. I use AI as my thought partner every day. I'm training my team to do the same.

Every day, I form unbreakable habits of applied AI. I am showing my team, my industry, and myself what it means to embody the promise of generative AI—not in theory, but in practice.

I am just scratching the surface of what I will implement with Fifth Avenue Properties, and my work on using AI in my personal life.

This is what walking the talk looks like.