Better coffee starts at home
The Roastery Lab
A working laboratory, open to read. The science of the entire process — every step of it.
Pacific Roots Roastery Lab · powered internally by RoastOS
Pacific Roots roasts like a data science lab, teaches like a field guide, and builds tools like an AI-native analytics company.
We do the science many small roasters aren't equipped to do — then publish it. The dashboards aren't a feature of the coffee; they're the operating system behind it.
01
Roast Data Capture
Every batch is data: green lot, moisture, density, charge temp, turning point, yellow, first crack, development, drop, weight loss.
02
Roast Visualization
Bean temp, environment temp, and rate-of-rise as decision-ready charts — phase bands, event markers, reference overlays, deviation.
03
Experiment Engine
Roasting as controlled experiments, not vibes: hypothesis, control roast, one variable, cupped conclusion, production decision.
04
Sensory + Brew Loop
The curve is only half the truth — structured cupping (seven axes), days-off-roast, extraction and TDS close the loop back to the cup.
05
Governed AI
An analyst, never an autopilot: propose-only, every recommendation cites its data and states its confidence. Humans approve. Everything logs.
What are we researching?
Research
Open questions, hypotheses, and the experiments that will answer them — starting with the rest-window study, E-001.
What do we measure — and how?
Methods
Green to cup, every step's science and the numbers we hold it to. Sourced, dated, and honest about target vs measured.
What have we learned?
Field Notes
The cast-iron origin, home-roasting guides, and the craft notes behind the coffee — where the lab teaches.
What can you explore?
Open Instruments
Every roasting-science instrument, interactive, on Pacific Roots reference data — the same tools, from Field depth to Lab.
Where's the proof?
Evidence
Reference profiles, batch reports, and experiment findings — each labeled by what it is: target, measured, or modeled.
How does it all connect?
The Platform
One domain model, three depths, a device-local backend, and a governed analyst that proposes but never decides.
Where the work stands today
Pacific Roots is pre-launch, and the Lab is honest about it: the reference profiles and house targets are real design intent; the instruments that need thermocouples, color meters, or a shipped batch run on clearly-tagged sample data until that capture connects. Nothing here pretends a target is a measured batch. When the first roasts run, the same instruments fill with real numbers — and you'll see the evidence change from target to measured, in the open.