Who is this assistant for?

You train and nothing changes on shift. There is no when: the team lives inside service. Every new hire is trained differently depending on who is beside them. Someone calls in sick and nobody can cover.

Restaurant training dies for two reasons: it is designed around topics instead of behaviours, and it ends in an attendance sheet instead of a test. This assistant builds it backwards —first what the person must do in front of somebody for it to count as learned, with its numeric cut-off, and only then the content that leads there— and ties it to what costs money today: waste, returns, ticket time, cash-count variance, turnover. You pick the stage —first-days induction, role training, refresher, the annual plan or the role assessment— and the role, and it changes the starting point, the duration, who teaches and what evidence closes each block. And the assessment is no longer a sheet of paper: it comes out as an interactive test the team member answers on a phone, which grades itself and builds its infographic report. Because all five stages train the same team but are designed inside out from one another.

AI assistant · N5 Complete system

◆ Included in the annual plan🎓 Onboarding and Trainingv1.1.0 · 2026-09-24

📷 Works best with an AI that sees photos (invoices, dishes, screens) · optional

Training Programme Builder for Restaurants

You pick the stage and the role, and the assistant designs training backwards: the test you can fail first, the content after. You get the gap measured in cash, each block's test with its cut-off, a calendar that fits the shift, who teaches, and the business indicator that must move.

Depth

16/23

Your data

Compatible AIs

Steps to use it

Depth of the resulttap a level to see what you get

◀ Put out today's fireRun the year ▶

🏛️

N5Complete system

Several chained pieces you install and operate over time, with their review cadence. It changes how the business runs, not one decision.

◆ This assistant 110 of the 164 published assistants are at this level

Is this for you?

What it does and what you get

Restaurant training dies for two reasons: it is designed around topics instead of behaviours, and it ends in an attendance sheet instead of a test.
This assistant builds it backwards —first what the person must do in front of somebody for it to count as learned, with its numeric cut-off, and only then the content that leads there— and ties it to what costs money today: waste, returns, ticket time, cash-count variance, turnover.
You pick the stage —first-days induction, role training, refresher, the annual plan or the role assessment— and the role, and it changes the starting point, the duration, who teaches and what evidence closes each block.
And the assessment is no longer a sheet of paper: it comes out as an interactive test the team member answers on a phone, which grades itself and builds its infographic report.
Because all five stages train the same team but are designed inside out from one another.

What you end up with

The full programme for the chosen stage and role: five to eight gaps written as observable behaviours, with the signal that betrays them and their monthly cost; each block's test with what the person does, in front of whom, its numeric cut-off and what happens on a fail; the minimum content with its house source and what is still to be written; the calendar in sessions under 45 minutes that never touch the peak window, with total hours per person and their payroll cost; who teaches each block and why; the team coverage matrix and the evidence table; the business indicator with its baseline and deadline; and the verdict on whether the programme actually fits. If you choose «role assessment»: the role's interactive test on the Operating Document kit template —12 to 20 questions on real shift situations, with competencies, weights, pass marks, critical questions and a practical test—, which grades itself and delivers the report with the score, the result per competency, what failed with its correct answer and source, the refresher plan with a date and the signatures.

EXAMPLE This is what you getAn example built with Sazón de Origen, the house's fictional restaurant.Not your data: your result is built with YOUR restaurant's.See the full example →

Sample result

Sazón de OrigenFictional contemporary Colombian restaurant in Medellín, 2 locations, 6 years in business. Every sample result in the library comes from this same case. Fictional restaurant

The owner asked this: «Every server who joins is trained by whoever is standing next to them and leaves within two months. I have 14 floor staff across two locations, ticket time has drifted to 14 minutes at the peak and returns are up. I want the server role training, not a customer-service course.»

Six behaviours, not six topics. The right-hand column decides what gets taught first.

Missing behaviour	Signal that betrays it today	Monthly cost
Takes the drink order before the food order	Drink arrives with the main on 6 in 10 tables	~1.2 M COP of unsold drinks
Enters the ticket with modifications written	9 returns/week for «that is not what I ordered»	~1.6 M COP in refires and waste
Offers the starter of the day by name	Starters on 11 % of tables	~2.1 M COP of uncaptured check
Checks the plate before carrying it	4 plates/week leave without their side	Return + kitchen time
Takes a complaint without calling the captain	The captain handles 8 complaints/week that do not need him	Captain off the pass 40 min/day
Closes the table without leaving the bill waiting	Bill wait > 6 min on 30 %	Table rotation lost at the peak

Illustrative figures on 14 floor staff and 100 M COP monthly sales. Note what is NOT here: «service attitude», «teamwork», «vocation». They cannot be tested, so they cannot be taught or demanded.

Block	What the person does	In front of whom	Pass mark	If they fail
B1 · The table sequence	Works 3 real tables start to finish	Captain, with a stopwatch	Drink < 5 min on all 3; ticket written without errors	Repeats shadowed on the next shift
B2 · The menu by heart	Describes 8 dishes without looking and names 2 allergens in each	Head chef	7 of 8 correct	Back to B2, does not move to B3
B3 · The suggestion	Offers starter and dessert on 5 consecutive tables, by name	Captain	At least 2 of 5 accept	One shadowed shift and repeat
B4 · The complaint	Resolves 2 simulated and 1 real complaint without calling anyone	Captain	Both within their authority, no escalation	The authority matrix is reviewed first

No test says «understands» or «knows». Every one says what they do, in front of whom, and what number passes.

Five sessions, none over 40 minutes, none inside 12:00–14:30.

Session	Minutes	Time	Mode	Who teaches
B2 · The menu	40	15:30	Session with a 4-dish tasting	Head chef
B1 · Sequence (theory)	25	15:30 day 2	Laminated sheet + walkthrough on an empty floor	Captain
B1 · Sequence (practice)	—	2 shifts	Shadowed shift	Reference server
B3 · Suggestion	30	16:00 day 5	Script + paired rehearsal	Captain
B4 · Complaint	35	16:00 day 8	Real cases from the last week	Captain

The cost, written before approval: 2.2 session hours + 2 shadowed shifts per person. Across 14 people that is ~31 session hours and 28 shifts — and the teacher's hours are paid too.

It comes out as a file answered on a phone (`ASSESSMENT_SERVER.html`), not as a text quiz. Summary of what it holds:

Competency	Questions	Weight	Pass mark	Source
Order taking and ticket	4	3	80 %	MC-SAL-01
Menu and suggestive selling	2	2	70 %	GUI-SAL-03
Allergens and food safety	2 · critical	3	100 %	PRT-GEN-02
Bill and table close	2	2	80 %	GUI-SAL-05

Critical question. A guest says they are allergic to peanuts. What do you do? → Mark it on the ticket as ALLERGY, tell the head chef out loud and confirm the dish against the allergen table.

The report comes out on its own at the end: overall score in a ring, each competency against its pass mark, what failed with the correct answer and its source, the verdict —PASSED, REFRESHER or NOT PASSED (one failed critical item means no pass)—, the retest date and both signatures.

Illustrative example generated with a fictional restaurant. Not a promise of results, not a client case.

This is 4 of 6 parts. Behind the wall are the two pieces that turn the programme into something you can run on Monday. Content and logistics: each block in short bullets with the house source it comes from —standard recipe, server operating manual, complaint protocol— and flagged in red what has to be written because it does not exist yet; who teaches each block, with the reason it is that person and what they must prepare; the evidence table with what gets signed, where it lives and how long it is kept; and the team coverage matrix, person by block, to see at a glance who can cover for whom. And the follow-up: the business indicators with today's baseline, their target and their thirty- and ninety-day deadlines, the random test repeated monthly with a different person, the half-yearly review of the whole programme against the menu and the team you have by then, and the verdict on whether this actually fits the hours that exist.

See the full example — free account

Works with these AIs

Claude — the best for this assistant The best fit here: it does not let a block in without a test and does not write tests that say «understands». Demand the numeric pass mark on every one.

How to use it

Pick the stage and the role at the top. The assistant changes the starting point, the duration and the test, not just the label.

Fill in My Restaurant with team size, peak windows and your yardsticks: the gap measured in cash and the payroll-hour cost come from there.

Before you start, have two figures that hurt today at hand —waste, returns, ticket time, turnover—: training is designed against those and not against topics.

Copy the whole assistant and paste it into your AI of choice. Answer its questions one at a time.

If you chose «role assessment», download the Operating Document kit and open plantilla_evaluacion_del_cargo.html: if your AI makes files it hands you the finished assessment; if not, paste the JSON it gives you into «Paste assessment». The person answers on a phone and the report comes out on its own, ready to print and sign.

Run the first block with three people, not the whole team: if the test cannot be applied exactly as written, fix it there.

Measure the gap indicator at thirty and ninety days. If none moved, you trained on what did not hurt.

Which of your data it uses

of 23

This assistant works with 16 of the 23 fields in «My Restaurant», among them Restaurant name, Business type and Cuisine or concept. It does not ask out of curiosity: these are what make the answer speak about YOUR scale instead of an industry average — the same calculation on a business of another size returns a number you cannot decide with. You fill them once and they apply across the library, so the second assistant you open already starts with them in place. Whatever you leave blank, the assistant asks for one thing at a time instead of inventing it, and says so before answering, so you know what it is working from.

Restaurant nameBusiness typeCuisine or conceptCity and countryNumber of locationsYears in businessTeam sizeStrong and slow daypartsActive sales channelsBest-selling dishesWhat makes you differentCommunication toneAverage check (with currency)Approximate food cost (%)Approximate labor cost (%)Main goal this year