Waylight · crew planning

The day, already
in order.

Waylight turns a yard's worth of moving work into one legible day — every crew, every slot, every handoff accounted for before the gate opens.

Runs 214 yards on the morning shift

Today · North yard

6262hours scheduled

  • Frame inspection — Dock 2

    07:30 · Crew A

    Done

  • Hull coating, second pass

    10:15 · Crew C

    Live

  • Rig load test

    13:00 · Crew A

    Queued

  • Handover walkthrough

    16:30 · All crews

    Queued

Waylight · who runs on it

Two hundred and fourteen yards, by trade.

The named ones agreed to be named. The counts are everyone, including the ones who would rather not appear on a marketing page.

  • Deep-water terminals

    6161yards

    • NORTH BASIN
    • KETTLE POINT
    • ALDER QUAY
    • GRAVEL HEAD

  • Inland and river ports

    8888yards

    • MILLRACE
    • TWO BRIDGES
    • LOWER FORD
    • STONE LOCK

  • Shipyards and dry dock

    3434yards

    • DRY DOCK 2
    • HALYARD WORKS
    • CANTRELL

  • Cold chain

    3131yards

    • COLD STORE CO
    • NORTHWIND
    • PALE HARBOUR

Waylight · build infrastructure

The parts that make the pipeline fast.

Ferrite caches what it can prove, parallelises what it can't, and shows its work either way.

Throughput, this quarter

+157%

Builds per hour across the fleet — cache hits excluded, so the line only moves when real work gets faster.

Builds per hour: 12 points, from 42/hr to 108/hr

Median wait

3838seconds

From push to first log line, p50 across every queue.

Proven caching

A step is only skipped when its inputs hash identical — never on a guess, never on a clock.

content-addressed

Every branch, first-class

Preview environments per branch with the same cache lineage as main — a fork costs a pointer, not a rebuild. Merges inherit everything the branch already proved.

Waylight · one day, worked

What a Tuesday does by itself.

No demo flow, no numbered steps — just the four moments where the product earns the day.

  1. 05:55

    The board re-plans itself

    plan r2, posted

    Overnight changes — a late barge, a crane hold — are already folded into the morning before anyone arrives.

  2. 06:40

    Gates open on the same page

    0 conflicting copies

    Every crew reads the same rows at the gate. The briefing is a glance, not a meeting.

  3. 11:20

    A wobble propagates once

    1 change, 2 boards

    One reshuffle updates both yards that share the crane. Nobody re-keys anything anywhere.

  4. 17:05

    The day files itself

    day 214, on the record

    The shift closes into the ledger as it happened — the review will quote rows, not memory.

Waylight · the same morning

06:40, before and after.

The radio morning

  1. 06:40Yesterday's plan, already wrong
  2. 07:10The crane argument, on the radio
  3. 08:00Crew B waits; nobody knows for what
  4. 09:30The plan forks into three notebooks

The ledger morning

  1. 06:40The board re-planned around the crane overnight
  2. 07:10Both crews read the same row
  3. 08:00Crew B is on the second slot already
  4. 09:30One record; the notebooks stay in pockets

Same yard, same weather, same crane. Different morning.

THE MORNING, MEASURED

214214
yards on the morning shift
1,1801,180+
crews planned daily
3131k
handoffs a week
99h
recovered per yard, weekly

Counted across every live yard, updated at shift close.

Waylight · one yard, at length

Eleven years of whiteboard, and the six weeks after.

North Basin Terminal — Bulk and project cargo · four yards · 210 crew

North basin ran on a whiteboard and two radios for eleven years. It worked, in the sense that the ships left. What it cost was the first ninety minutes of every morning, which went to establishing what had actually happened overnight before anyone could decide what to do about it.

The first thing that changed was not the planning. It was that the overnight events — a late barge, a crane held for inspection, a crew short — arrived as rows instead of as rumours. By the time the gate opened, the argument about what was true had already been settled by the record.

The morning stopped being an investigation and became a decision.
M. Aldana — Yard lead, north basin

The planning changed second, and more slowly. Crews trusted the board about six weeks in, which is roughly when the third or fourth reshuffle propagated correctly without anyone re-keying it. After that the whiteboard stayed up for a month out of superstition, then came down.

Waylight · pricing

Pay for the floor you actually run.

Every tier includes the full instrument set. The difference is scale, history, and who answers when you call.

Field

For a single bench getting its footing.

$00/ seat / month

 

  • One bench
  • 14-day history
  • Community recipes
Most chosen

Crew

For teams that ship every week.

$1919/ seat / month

billed annually

  • Unlimited benches
  • Full history, forever
  • Shared recipe lineage
  • Priority queue

Works

For the whole floor, with controls.

$5252/ seat / month

billed annually

  • Everything in Crew
  • SSO and audit export
  • Dedicated capacity
  • Support with a name

Waylight · plainly

The last four questions, answered whole.

No drawers, no search. These are the ones people actually ask on calls, answered the way we answer them there.

What are you actually selling?
A morning that runs on a shared record instead of radio and memory. The software is the medium; the product is the argument you stop having.
Why would this work where the last tool didn't?
Because it starts at the gate, not the office. The board is built for gloves and glare, and the office views are derived from what the yard already touched.
What breaks if we stop paying?
Nothing retroactively. Exports keep working, history stays yours, and the free tier keeps a single yard running. Leaving is a workflow, not a hostage exchange.
What is this bad at?
Long-horizon planning. Waylight is a morning instrument — quarters and budgets live better in the tools you already have, and we integrate instead of pretending.

Two ways in. Both real.

SELF-SERVE

Walk in

Open a bench now and find your own way around. Most teams are running before lunch.

  • No card, no call
  • Import whenever
  • Leave with your data

GUIDED

Be walked through

Thirty minutes with someone from the field team, on your data, with your questions.

  • A person, not a demo reel
  • Your data, live
  • No follow-up sequence