David MartinAI knowledge systems & contentGet in touch
AI systems

I build the tools behind the work.

Running a whole marketing function alone only works with good systems. These are the ones I designed, built and use every day. People review everything before it ships.

These are in-house systems, so I describe what they do rather than show their screens. I'm happy to walk through them live in an interview.

Problems I solve

Where teams lose time, and how I get it back.

These are the bottlenecks leaders describe most often. Each system below exists because I hit one of them.

The problemEvery proposal starts from a blank page, and setup eats days.What I do

A structured AI workspace that drafts from approved answers and flags gaps for experts, so the team spends its time on strategy.

The problemAI pilots impress in the demo, then stall.What I do

Knowledge bases, clear guardrails and human review built in from day one, so the tools survive contact with real work.

The problemContent approvals are buried in email and nobody knows what is waiting.What I do

One workflow with clear statuses, from draft to published, that everyone can see.

The problemOpportunities slip past because nobody has time to check every portal.What I do

A daily digest that screens the sources for you and puts the right bids in front of leadership.

Built with Claude on SharePoint

Proposal Guru

The problem
Every new RFP meant 1 to 2 days of setup and weeks of drafting across many authors.
What I built
A structured AI workspace with house style, templates, a tracker and one folder per bid. It flags gaps for experts and keeps its own status report current.
The result
A new RFP becomes a folder, branded templates and an 80 to 90 percent complete first draft in about 30 minutes.
  1. New RFP
  2. Folders and templates
  3. AI first draft
  4. Expert review
  5. Submission
Built with Claude on SharePoint

Social Media Register

The problem
Post approvals lived in email threads, and nobody could see what was waiting on whom.
What I built
One system that drafts post copy and graphics, then moves every post through review with a checklist and a clear status.
The result
Every company post now goes through it, from first draft to published.
  1. Draft
  2. Awaiting approval
  3. Changes required
  4. Approved
  5. Scheduled
  6. Published
Microsoft Copilot Studio

AI agents for the whole company

A bid response assistant, a marketing expert, a Google Ads expert, a cybersecurity advisor, a sales sequence builder and a 3D animation-style image prompt generator.

One bid agent produced a 35-page RFP response live in a leadership session.

Weekday digest

Bid Radar

A morning digest of new public-sector opportunities for the 11-person leadership and sales group, screening BC Bid, CivicInfo BC and 16 buyer portals. Revised four times in its first three weeks from reader feedback.

Alongside it, I consolidated 32 years of bid history, more than 1,700 pursuits, into one traceable record.

Everyday tools

Claude projects I use every day

One-pagers, video scripts, blog posts, business cards and sector outreach campaigns. The company's social posts are written with a Claude project I built on a shared knowledge base, and a companion project turns each post into a full blog article in the house style.

Teaching

Helping people work with AI.

I present at the company's AI pilot program and train colleagues on the systems I build. This is the longer of my two ELIZA effect videos.

The ELIZA Effect: How Human Habits Quietly Break AI Collaboration

How human habits quietly break AI collaboration

About five minutes. Built as a presentation, then turned into a narrated video with Microsoft 365 Copilot.

  • PowerPoint
  • Microsoft 365 Copilot

Produced for MYRA Systems Corp.

Recognise one of these bottlenecks?

I'm happy to walk you through the live systems and talk about what the same approach could do for your team.

Start a conversation →