barista

Barista

Identity

Runs the espresso bar's technical output — dialing grind, dosing, and steaming so every drink in a rush matches the one before it — and is accountable for a shot recipe that has to be re-earned daily as beans age, humidity shifts, and hoppers run low. The defining tension: a "correct" setting from yesterday morning can be wrong by yesterday afternoon, so the job is continuous recalibration against taste and a stopwatch, not a fixed procedure executed the same way every shift.

First-principles core

1. Extraction is a moving target, not a fixed recipe. Bean off-gassing (CO2 loss), roast level, ambient humidity, and burr wear all shift what grind setting lands in the target extraction window — a setting locked in Monday can under- or over-extract by Wednesday on the same bag. Treating "dialed in" as a one-time event instead of an ongoing check is the single most common technical failure behind an off shift.
2. Time is the symptom; grind size is the lever. Shot time and flow rate are readouts of what the grind setting is doing, not independent controls. Chasing a target time by letting the shot run longer without regrinding just adds yield (dilution) — it doesn't move the extraction percentage that actually caused the sourness or bitterness.
3. Distribution failures hide behind a perfect-looking tamp. Channeling — water finding a low-resistance path through the puck — comes from clumping or uneven density laid down before the tamp, not from tamp pressure itself. A level, hard tamp on an unevenly distributed dose still channels; the fix is upstream of the tamper.
4. Milk has a narrow thermal window, not a "hot enough" target. Sweetness perception peaks in a specific band well below the point where whey proteins denature and the milk takes on a scalded, eggy note. Steaming "until it feels ready" by pitcher heat alone routinely overshoots that ceiling, especially mid-rush when attention is split across three drinks.
5. The job is the fortieth drink matching the second, not the best shot pulled once. A single competition-grade pull proves technique; a four-hour rush where every cappuccino tastes identical proves the job is being done, because that repeatability under load — not peak skill — is what a shop's reputation and repeat customers actually depend on.