[ past work ]

The work, in full.

Five engagements written up the way we actually thought about them. What was broken, what we built, how it shipped, and the number that moved. Skim the headline if you just need proof. Read the architecture if you want to judge it.

small business / startup ai

Turning a lean operator into an AI-native business without hiring a full AI team

30 days
first workflow
Lean
team size
1
proof asset
read the case
retail ai / pricing

Building a price and promotion system that moved margin and demand together

+5%
net margin
+10%
demand
4
decision scope
read the case
growth ai / mmm

Turning marketing mix modeling into an executive planning system

MMM
model family
Digital twin
planning loop
Spend allocation
use case
read the case
financial services / operations ml

Propensity and throughput models for a lending operation that needed cleaner signal

+50%
propensity lift
~2x
qualified leads
31 -> 29
cycle time
read the case
agent systems / evals

A multi-agent orchestration layer with evals before autonomy

-30%
token usage
RAGAS
eval layer
Tool calls
risk focus
read the case
[ start here ]

Bring us the job you repeat.

[ modalis — working session ]
Twenty-five minutes on one job from your business.

Bring the workflow you run through a frontier model on repeat, the one whose bill grows with your volume. We’ll tell you straight whether a small local model can clear the same bar, and what a run costs on hardware you own.

What we cover
  • Whether a small local model can match the frontier model on your job
  • How we would measure that its answers are correct
  • What hardware it needs, what a run costs, and a rough timeline
25 minutes
Google Meet
America/Toronto
email us to schedule
[ book directly ]

Tell us about the job.

Send a short note describing the work you want handled and the systems it touches. We will reply with times for a 25-minute session.

email to book