Free · 3-5 days · No obligation
AI Opportunity Audit
We map your operations, find the highest-ROI automations, and hand you a ranked plan with payback math. Yours to keep, whoever builds it.
No obligation. No follow-up sequence.
AI content engine
A content engine that publishes every week, in your voice
Drafting, repurposing and scheduling from your actual expertise. Humans approve everything. 5x ROI in 30 days. Or we work for free.
Teams we build for
- Hoyes Michalos
- Nurse Next Door
- Fedi
- UBC Sauder
- Merchant House Capital
- Picton Investments
- Campbell Froh May & Rice LLP
- Barnakl
- Hungerford
- Breez
Teams we build for
- Hoyes Michalos
- Nurse Next Door
- Fedi
- UBC Sauder
- Merchant House Capital
- Picton Investments
- Campbell Froh May & Rice LLP
- Barnakl
- Hungerford
- Breez
Teams we build for
- Hoyes Michalos
- Nurse Next Door
- Fedi
- UBC Sauder
- Merchant House Capital
- Picton Investments
- Campbell Froh May & Rice LLP
- Barnakl
- Hungerford
- Breez
The problems this solves
01
Publishing is feast or famine.
Content goes out when someone has time. Someone never has time.
02
The founder is the bottleneck.
Every post waits on the one person whose calendar is already full.
03
Generic AI output embarrasses the brand.
You tried the tools. Everything came out sounding like everyone else.
04
Repurposing never happens.
The webinar was great. It became zero posts, zero clips, zero emails.
05
SEO rewards consistency you can't sustain.
Rankings need a cadence. Manual production can't hold one.
What we build
01 · Voice capture
Your real writing, calls and positioning distilled into a voice the system writes in. Not a style dropdown.
02 · Topic pipeline
Topics sourced from your actual expertise: client conversations, internal docs, the questions you answer every week.
03 · Drafting
Long-form drafts written in the captured voice, grounded in your material.
04 · Repurposing
One piece becomes many: post to newsletter to social, each fitted to its channel.
05 · Review queue
Nothing publishes without a human approving it. Your team edits in minutes instead of writing for hours.
06 · Scheduling and distribution
Approved content ships on cadence, every week, without anyone remembering to do it.
Nothing publishes without a human approving it. Newsletters are built CASL-clean: consent tracked, unsubscribe honored.
What's included
- Voice capture and a directive file your team can edit in plain text
- The full pipeline wired to your channels
- Review workflow your team actually uses
- Training, monitoring and iteration after launch. Cancel anytime
Proof
BreezNamed with consent
Lead generation · Live
Breez takes prospect research from hours to minutes with an outbound engine
Breez spends minutes per researched prospect instead of hours, with 40+ personalized data points per proposal. The outbound engine behind it, in full.
Minutes per researched prospect, was hours
A healthcare booking marketplaceDescribed, not named
Healthcare · Live
A healthcare booking marketplace publishes 9,437 pages across three provinces with a local-SEO engine
A booking marketplace taken over from a prior vendor, then a local-SEO engine on top. 9,437 pages live in three provinces, first region live 8 days in.
9,437Build artefact
Weeks, not quarters.
How the full engagement works01 Discovery
WEEK 0
02 Prototype
WEEKS 1-3
03 Deploy & train
WEEKS 4-8
04 Run & improve
WEEK 9+
How we build it
Step 01
Capture the voice before generating anything.
We index what you have already published and said. A model that has not read you writes like every other company in your category.
Step 02
Draft against a graded standard.
Output is scored against pieces you stood behind, for voice and for factual accuracy, before the cadence starts.
Step 03
Then hold a cadence.
Repurposing and scheduling run once quality holds. Publishing on a schedule is worthless if the drafts need a rewrite every time.
What we will not automate
Publishing without a person, and any factual claim that does not trace back to a source. We do not generate volume for its own sake.
How it is built
See the full stackRetrieval
Your own published material, transcripts and positioning are indexed so drafts sound like you and cite your actual claims. This is the difference between content and generic filler.
- pgvector
- Full-text BM25
- Voice corpus
Agents and orchestration
Topic research, drafting, repurposing and scheduling run as stages in one graph you can open and change. Every stage leaves its output for review rather than passing silently downstream.
- agent-worker
- Workflow builder
- Proposal queue
Evaluation
Drafts are graded against pieces you have already published and stood behind, scored for voice as well as accuracy. Every factual claim is checked back to a source before publication.
- Eval graders
- Voice grading
- Claim checking
Models
Frontier models for drafting in the captured voice. Embedding models to keep the piece grounded in your material rather than in what the model remembers about your industry.
- Frontier models, one gateway, routed per task
- Embedding models via the same gateway
- Grounding checks
Data boundary
Your material, positioning and unpublished drafts stay in your accounts under no-training terms. Nothing you have not published becomes anyone else's training data.
- Supabase row-level security
- No-training API terms
- Draft isolation
Named tools in this build: Claude · Workflow builder · Supabase · Vercel + Next.js
What that means in practice
Retrieval and RAG
Connecting AI to your actual documents so it answers from your knowledge, accurately and with citations, instead of making things up.
Where we stop. A dedicated vector database is justified by scale, not by default. Most of RAG quality is won or lost in chunking and indexing strategy, not in the model choice, and we have walked clients back from RAG to plain search when that was the honest answer.
Agents and orchestration
AI that does the work instead of just answering: looks things up, calls your systems, completes multi-step tasks, and knows when to hand off to a human.
Where we stop. Multi-agent swarms are oversold; most jobs need one well-guarded loop. If a cron job and a script solve it, that is what we build, because 90 percent per-step accuracy compounds to 59 percent over five chained steps and no framework changes that arithmetic.
Evaluation and observability
How we prove the AI actually works: measured, monitored and regression-tested like real software, not vibes.
Where we stop. There is no engagement where we skip this. The honest variable is depth: a document pipeline gets faithfulness and extraction suites, an outbound agent gets human review sampling, a classifier gets a held-out test set. We size the harness to the risk, never to zero.
Questions? Straight answers.
Also built here
Three ways in. No long discovery.
Paid · Fixed scope
First Build
One high-ROI system, built on your real data and deployed in your stack, with your team trained to run it. Fixed scope. Quoted after the audit. Covered by the 5x ROI guarantee.
Scope my first build
Paid up front. Cancel anytime after.
Free · 15 min
Intro Call
Fifteen minutes with James, not a sales rep. Bring your worst bottleneck, leave with a straight answer.
Book 15 minutes
No pitch deck.
Every first build is covered: 5x ROI in 30 days. Or we work for free. Read the full terms
Start free. Know your number in five days.
A 3 to 5 day audit of your operations, ending in a plan with the ROI math attached. No obligation.