Applied AI-Powered Bioinformatics Tools

Instrument Schedule Visualizer

Last reviewed

Beginner

What you'll learn

~40 min
  • Ask an AI tool, in plan mode, for a page that reads a week's booking export and your trained-user list (who has been trained on which instrument) and shows each day's bookings with what is wrong with them
  • Answer its questions with your facility's rules: what counts as a clash, what happens to cancelled bookings and overnight runs, and what each booking must have
  • Check the page's list of problems against clashes, bookings over maintenance and untrained users worked out for the practice files
  • See what the page cannot decide when an instrument goes down mid-week, and whose calls those are

The story, and what you will end up with

It is Thursday afternoon, 12 November 2026. You manage bookings for your core facility’s shared instruments, and every Thursday you look over next week’s bookings, Monday 16 to Sunday 22 November, so that anything wrong can be sorted out with the labs before the week starts. This week there is more to look at than usual. Staff entered some bookings by hand from emails, and the booking system lets staff place a booking over someone else’s. And yesterday the confocal’s service engineer moved their preventive-maintenance visit to Tuesday morning, after one lab had already booked that time. You told that lab yesterday; their booking stays in the system until they choose a new time.

You have exported next week’s bookings from the booking system, and you keep a trained-user list: who has been trained on which instrument. Reading forty rows by eye for overlaps is slow, and an overnight run that clashes with the next morning is easy to miss.

By the end of this lesson you will have a schedule page on your own computer: a page in your browser that reads a week’s booking export and your trained-user list, shows each day’s bookings instrument by instrument, and lists what is wrong with them. It changes no booking and sends nothing anywhere. You build it by asking your AI tool in plain English, and you check it against problems this lesson has worked out for the practice files.

🔍Words used in this lessonclick to expand
  • Booking export: the file of bookings your booking system writes out when you ask for a report, one row per booking.
  • Clash (or overlap): two booked sessions on one instrument whose times overlap, by any amount. In this lesson, a session that starts at the minute the previous one ends (back to back) is not a clash.
  • Maintenance block: time booked on an instrument for maintenance, such as a service engineer’s visit or a clean, so that no one else books it.
  • Trained-user list: the facility’s record of who has been trained on which instrument, and so may book it.
  • Overnight run: a session that starts one day and ends the next, such as a time-lapse that images cells through the night.

Before you start

First, if your AI tool is still open, leave it: type /exit and press Enter (if you use Antigravity, press Ctrl+D with nothing typed instead), and wait until you see the $ or % prompt. The next line goes there, in the terminal, not in the AI tool. It makes a folder for this lesson beside your practice folder and goes into it:

Terminal window
mkdir -p ~/bio-workbench/22-instrument-schedule && cd ~/bio-workbench/22-instrument-schedule

Start your AI tool with the start line you kept from the setup lesson, the one with approval prompts off. Then ask it to set the folder up:

Set this folder up so we can undo any change you make, download https://learn.intellimetrics.net/starters/biotech/22/schedule.zip and unpack it into a data folder, and save a checkpoint.

Look for it saying three things: the folder is set up with Git (the program that saves snapshots, from setup), two files are in the data folder (bookings-2026-11-16.csv and trained-users.csv), and a first snapshot, your checkpoint, is saved. Later you can ask whether the files have changed since then. If it does not mention one of them, ask: “Did you save a checkpoint with the files in it?” It may also keep the zip file in the data folder, which is fine; if it says it kept one somewhere outside this folder, ask it to delete that copy. If the download does not work, download schedule.zip in your browser, check where it went (the browser’s list of downloads shows the folder), and tell your AI tool, for example: “schedule.zip is in my Downloads folder. Unpack that one file into a data folder here, and save a checkpoint.” In Windows PowerShell or a desktop app, “my Downloads folder” works as is; only say “my Windows Downloads folder” if you use Ubuntu (WSL) on Windows, which keeps its own Downloads folder apart from Windows.

The box below says what is in the practice files; later steps refer to them.

ℹWhat is in schedule.zipclick to expand

Two files, each of which opens in a spreadsheet program.

bookings-2026-11-16.csv, next week’s booking export: 37 rows, from Monday 16 to Saturday 21 November (nothing is booked on Sunday). Each row has a Booking ID, the Instrument, its Kind (Session or Maintenance), the User, their Group, the Account the session is billed to, its Start and End (such as 2026-11-18 18:00), its Status (Booked or Cancelled) and a Note. The rows are in order of start time, so the booking IDs are not in order.

InstrumentWhat it isRows
CONF-1Confocal microscope11 sessions (one cancelled), 1 maintenance
FLOW-1Flow cytometry analyser9 sessions, 1 maintenance
QPCR-1Real-time PCR machine10 sessions
LIVE-1Live-cell imaging microscope, booked for overnight time-lapse runs5 sessions

Users are U01 to U14, in groups LAB-A to LAB-F, each group with one account (ACC-2031 to ACC-2089). S01 is facility staff, in the group FACILITY. The two maintenance rows are booked by S01 and carry no account:

RowInstrumentWhenNote
M01CONF-1Tuesday 17 November, 08:00 to 12:00Service engineer visit, preventive maintenance
M02FLOW-1Thursday 19 November, 08:00 to 09:00Weekly clean and QC

trained-users.csv, the trained-user list: one row for each person and instrument they are trained on, with the date. S01 is on it for all four instruments; U01 to U13 for one or two each. U14 is not on it.

Everything in these files is made up for this course: the facility, its instruments, users, groups and accounts. Each lesson in this track has its own practice files, with their own people and dates.

Switch to plan mode

Your AI tool is already open. Switch to plan mode now: Shift+Tab until it says plan, or /plan (the keys for each tool). In plan mode it usually reads your files, asks its questions and shows you a plan before it builds anything; the order can vary.

The conversation

Each message below is an example. Say it your own way.

Say what you want

Say what you do, which files to use, and what you want from them:

I manage bookings for our core facility's shared instruments. From the two files in data, build a small page that shows next week's bookings and what's wrong with them. Ask me what you need first.

Your AI tool usually reads both files and comes back with questions. It may also list problems it can already see.

Answer its questions

Answer each in a sentence. These are questions the tools often ask, with one way to answer each. Yours may ask them in a different order or show a list of choices:

If it asksYou might say
Should it flag overlaps, maintenance clashes, missing training and missing accounts?Yes, flag all four.
Is back to back fine, or do some instruments need a gap between users?Flag overlaps only; back-to-back can stay for now.
Does training expire?Use the trained list as it stands, no expiry.
Are evenings, nights and weekends allowed? Overnight runs on any instrument?Fine on every instrument. Show overnight runs on both days.
Is there a longest a session may be?No length limit for now.
Should it check accounts? Does maintenance need one?Flag a booking with no account; maintenance needs none.
Which week, and which time zone?16 to 22 November, our local time.
A calendar for the week, or a table?A calendar for the week with a list of problems.
A page just for you, or shared?One page for me, working without the internet, rebuilt from next week’s files.
Read-only, or should you be able to change bookings?Read-only. List problems only; I’ll sort them out with the labs.

If it asks several questions at once, answer the ones about what counts as a problem from your facility’s own procedures, a sentence each, and accept its suggestions only about how the page looks. If you cannot tell which kind a question is, treat it as a rule. If it asks something not listed here, answer it the same way, in a sentence.

These answers are this practice facility’s own rules. Check whether yours require a gap between users, for a rinse, a clean or a wash, or allow two users at once on an instrument, and whether they require a sign-off before anyone works evenings, overnight or alone; the practice files hold no such sign-off record, so this page cannot check it. Your rules are in your facility’s own procedures.

Read the plan

Next it usually shows you a plan: what it will build and how it will check it. Read it for six things:

  1. Every row of the export is read, and the maintenance rows and the cancelled booking are kept apart from booked sessions. A cancelled booking is shown but not checked.
  2. Clashes are found by time, across midnight too: an overnight run is checked against the next morning’s bookings, not only against bookings that start the same day. Back to back is not a clash.
  3. Each booked session is checked against the maintenance blocks, against the trained-user list for that instrument, and for an account.
  4. The page lists problems and nothing more. It does not move, cancel or rebook anything, and it does not choose who keeps a slot.
  5. It can be run again on next week’s files.
  6. Your two files are read and left as they are, and nothing is sent anywhere.

A first plan often leaves one of these out. Say so in a sentence, in your own words, for example:

Also check overnight runs against the next morning's bookings, not just against bookings that start the same day.

When the plan has all six, approve it: if your AI tool offers a choice to go ahead, pick it; otherwise say “Yes, go ahead and build it.” In Codex, press Shift+Tab first to leave plan mode. In Copilot, saying it brings up a choice to go ahead; pick that. Let it build. It may test what it built and tell you how to open it. When it has finished, ask it: “Did you install anything, and where?” Nothing, or something inside this lesson’s folder, is fine for these lessons. If it names anywhere else on your computer, such as a few scratch files of its own, ask it what they are for, and to remove them if you do not want them there.

Look at what it built

Ask it to open the page:

Open the schedule page in my browser.

The page is usually a single file in this lesson’s folder, which your browser can open by itself. If your AI tool gives you the file’s location instead of opening it, often an address beginning file://, copy it into your browser’s address bar and press Enter. If it starts the page at a local address, one ending in a number, and says the address is already in use, say: “Something else is using that address. Don’t stop it. Use a different address for my page.” If your AI tool does not take a message after opening the page, wait a minute. If it still does not, close the terminal window, open a new one, type this lesson’s folder line again, start your tool, and say: “Start the page again so it keeps running while we talk.” The page and your saved work are files in the folder, so nothing is lost. When you have finished with the page, ask: “Stop the schedule page if it is still running.”

Your page will be laid out in its own way. Check it for these:

  • It covers Monday 16 to Sunday 22 November: 34 booked sessions, 2 maintenance blocks and 1 cancelled booking. Some pages count the maintenance blocks among the bookings and show 36.
  • It lists three clashes. On CONF-1, B01 (LAB-A) and B02 (LAB-C) overlap on Monday from 11:00 to 12:00. On LIVE-1, the overnight run B32 (LAB-D, Wednesday 18:00 to Thursday 08:00) and B33 (LAB-B, from Thursday 07:30) overlap on Thursday from 07:30 to 08:00. On FLOW-1, B19 (LAB-B) and B20 (LAB-E) overlap on Friday from 14:30 to 15:00. If it lists only two, the overnight one is missing: ask it to check runs across midnight.
  • B06 and B07 on Wednesday are not a clash, because B06 is cancelled; B06 is shown as cancelled. B13 and B14 on Tuesday, and the FLOW-1 clean M02 and B17 on Thursday, meet at the minute and are not flagged.
  • B04 (LAB-C) is flagged against the engineer’s visit M01 on CONF-1, Tuesday from 11:00 to 12:00.
  • Two sessions are flagged for training: B09, because U11 is trained on FLOW-1 only and has booked CONF-1; and B18, because U14 is not on the trained-user list at all.
  • B23 (QPCR-1, Tuesday 09:00) is flagged for having no account.
  • The overnight runs B31, B32 and B34 are each shown on both days they cover.
  • When you ask “Have my two files changed since the checkpoint?”, it compares them with the checkpoint and says no.

That makes seven problems. A page may also add checks of its own, such as rows it cannot read; on these files they find nothing.

If something on the list is not true, tell your AI tool in a sentence what you see and what you expected, for example: “B06 is flagged as a clash with B07. B06 is cancelled, so leave it out of the checks.”

Notice what the page cannot decide

The page tells you what is wrong. Now suppose FLOW-1 fails on Wednesday at noon and the engineer cannot come before Friday. Ask the following, including its last question, which is yours to answer, to see what your AI tool does with it:

Make a copy of the bookings file in this folder. In the copy only, mark FLOW-1 out of service from Wednesday 12:00 to the end of Friday. Then show me which bookings that touches, and who should move.

Before you read on: what do you expect it to say about who should move?

It should find five booked sessions inside that time: B16 (LAB-F, Wednesday 14:00), B17 (LAB-A, Thursday 09:00), B18 (LAB-F, Friday 09:00), B19 (LAB-B, Friday 13:00) and B20 (LAB-E, Friday 14:30), and the Thursday clean, M02. It will probably say that all five need to move, and that B15 (LAB-E, Wednesday 09:00 to 12:00) is not affected because it ends as the outage starts.

That is as far as the files go. They say who booked which hours. They do not say whose samples can wait and whose cannot, which lab has a deadline, whether a partner facility could take urgent samples, or when FLOW-1 will be back: check whether your procedures have a repaired instrument checked before it is released, and if they do, give no lab a date until it has been. Nor do they show that B15’s session ended at the moment the instrument failed, so that user’s results and samples are worth checking with them; they may be the person who reported the fault. B18 is a training question before it is a scheduling one: whether a user who is not on the trained list may run at all, and with whom, is set by your facility’s training policy, and is not something the schedule or the lab’s wish settles. The same policy decides whether that session becomes one with staff present, a training session, or is cancelled. U14 may be a new user waiting for training or someone booking under another person’s login, and either needs following up.

So the five users hear from you today, not at next Thursday’s review. Who gets the first slots back follows your facility’s booking policy, applied by you with the labs.

A schedule check lists clashes, bookings over maintenance and users not on the trained list. Who keeps a slot, what moves when an instrument goes down, and whether a flagged session goes ahead are decided under your facility’s policies, by its manager, with the labs.

When you have looked, remove the copy: “Delete that copy; I’m done with it.”

One real run

Here is part of what one AI tool said after the experiment above, with Claude Code on 11 November 2026. Yours will word it differently; check your page against the list above, not against this.

- **M02**, the weekly clean and QC on Thu 08:00–09:00, also falls
inside the outage. You'll want to decide whether to cancel it or
move it after the repair.
- **B15** (U10, Wed 09:00–12:00) ends exactly when the outage starts,
so it isn't affected.

The facility, instruments and bookings it talks about are made up for this course.

It counted B15 by the clock, which is all a file can do. That session is the one to ask about first. And the Thursday clean moves to after the repaired instrument has been checked, a date no file holds.

How real facilities differ

  • Check whether your booking system keeps maintenance in a calendar of its own rather than as rows in the bookings export. If it does, the page needs both.
  • Check whether your booking system refuses a booking that overlaps another. If it does and still shows a clash, find out how it got there: a staff override, a booking imported in bulk, or maintenance added after users had booked. If maintenance was added after a lab booked, check what notice your procedures say that lab is owed, and when.
  • Check whether your facility has a written booking policy that sets who keeps a contested slot: who booked first, cancellation deadlines, limits on busy hours for each group. Check too whether your export shows when each booking was made and last changed; this one does not.
  • Check whether your training list carries more than this one: training that lapses after a time without use, sign-off for evening and overnight work or for working alone, laser safety for a confocal, approval to run unfixed or biohazardous samples on a flow analyser. Your facility’s procedures say which apply.
  • Check which statuses and kinds your export carries beyond Booked and Cancelled, such as pending approval, held, no-show, completed, late cancellation, training, assisted or staff-run service. Ask your AI tool to show any it does not recognise rather than drop them.
  • Bookings are not use. Check whether what an instrument really ran, and what is billed, comes from its login or usage records at your facility; those can show no-shows and overruns that bookings hide.
  • Bookings change after you export them. Run the check again on Monday morning, and whenever the bookings change.
  • Where the clocks change, the night they change, an overnight run’s length on the clock and its real length differ by an hour. Check how your export writes times that week.

Keep going

Small changes do not need plan mode. Say each in a sentence, then look at the page again:

  1. Suppose this practice facility now sets a weekday limit of 4 hours on CONF-1; it is an example, not a general limit. Say: “Also flag any CONF-1 session over 4 hours on a weekday.” On the practice files it flags B10, Friday 10:00 to 15:00, five hours.
  2. “Mark sessions that are back to back on one instrument, so I can check the handover.” On the practice files that is B13 and B14 on Tuesday at 12:00, and your page may also mark B17 straight after the Thursday clean.
  3. “Save a checkpoint.” Your AI tool saves a snapshot you can return to.

When you want to check your real bookings, use the finished page. It runs on your real files by itself, with your AI tool closed, so your AI tool never sees them. Nothing confidential or unpublished, and nothing about people or about animals under a protocol, goes to your AI tool or into this course folder. Your booking system’s export will not be shaped like the practice file: its columns will have other names, and it will carry real names and accounts. Do these steps in order; the first five happen now, while your AI tool is open and this folder still holds only the practice files:

  1. Go through the data check, the questions in setup about what the data is, who decides how it is used, and which AI tool would see it. Booking records name people and accounts, and some facilities’ bookings concern human or animal work under a protocol. Go on only if your facility allows this use.

  2. Decide on a folder of its own for your real files, outside this course folder, for example ~/instrument-bookings. Nothing has to be in it yet. If the folder is not there, ask your AI tool to make it: “Make an empty folder called instrument-bookings in my home folder.”

  3. Ask your AI tool for blank files to fill in, for example: “Make me a blank bookings file and a blank trained list in ~/instrument-bookings, same columns, that my spreadsheet program can open. Put one made-up row in each, and show the page working on them.” Look for it building the page from the made-up rows. If your export’s columns have other names, tell your AI tool those names, with two rows you make up in the same shape, never real rows, and ask it to read files like that.

  4. Give the program your own facility’s booking rules in place of this practice facility’s: what counts as a clash (for example, whether yours needs a gap between users), who may book which instrument, and whether training expires. Say them in a sentence or two, then ask: “Make up one booking that breaks each of my rules, and show me the page flagging each one.” Check that each made-up booking is flagged for the rule it breaks.

  5. Ask your AI tool for the exact line, for example: “Write me the exact line to build the page myself from files in ~/instrument-bookings, with full paths.” Use your own folder’s name. If it offers to copy your real files into this folder instead, say no. Copy the line it gives you into a note. It will look something like this one, though yours may differ:

    Terminal window
    python3 /home/dana/bio-workbench/22-instrument-schedule/build_page.py --data /home/dana/instrument-bookings --out /home/dana/instrument-bookings/schedule.html

    Read it left to right. The page is written by a small program your AI tool made in this lesson’s folder; this line runs that program on your files and writes a fresh page. python3 runs a Python program. Next comes the program that builds the page. --data says which folder to read the bookings and trained list from, and --out where to write the page. /home/dana is the home folder of someone whose user name is dana; your own user name goes where dana is, and on a Mac the folders begin /Users/ instead of /home/. The line runs the program where it is, so leave this lesson’s folder in place. If you do not know your home folder’s full name, ask your AI tool: “What is the full name of my home folder?” Ask too how the files must be named and what form the times must take, because a spreadsheet program may rewrite 2026-11-23 09:00 as 23/11/2026 09:00 when it saves. If you are unsure about any part, ask your AI tool now, while it is still open.

  6. Leave your AI tool the way you did at the start of this lesson.

  7. Export next week’s bookings from your booking system and save your trained-user list into your folder, named as your note says. Paste the line from your note at the $ or % prompt and press Enter, then open the page it writes there in your browser. To come back to the lesson folder later, type this lesson’s folder line again, the line near the top of this page that begins with mkdir.

Try another case

Before the quiz, write two or three sentences for yourself. Next week the page shows no problems at all, but on Monday morning a lab tells you their overnight run on LIVE-1 was stopped at 07:00 by another user who had booked from 07:00. What would you check first, and what would you tell the lab?

A good answer checks the export the page was built from before blaming either user: whether the overnight booking was in it, how its end time was written, and whether either booking was made or changed after the export. It asks whether the page compares runs across midnight. And it tells the lab what happened from the records, while who gets time to repeat the run follows the facility’s booking policy.

Key takeaways

  • A short ask is enough. What you do, which files, what you want; your AI tool’s questions bring out your rules.
  • What counts as a problem comes from your facility’s rules, not from your AI tool’s suggestions.
  • Check a new page against problems you know before you trust it on a week you do not.
  • An overnight run is where a clash hides. The check has to compare times across midnight.
  • The page flags; you decide, under your facility’s policies, with the labs.
  • Practice data stays practice data: real bookings go to the finished page, with your AI tool closed, after the data check.
KNOWLEDGE CHECK

An overnight run on LIVE-1 is booked from Wednesday 20:00 to Thursday 08:00. Another lab has booked LIVE-1 on Thursday from 07:00 to 09:00. Your page lists no clash on LIVE-1. What is the most likely reason?

KNOWLEDGE CHECK

The page flags a booking because the user is not on the trained-user list for the confocal. Their group leader writes that the user has used confocals for years at another institution. What do you do?

What’s next

This is the last lesson in the Biotech Track. Each lesson builds a different tool, and each one can be done on its own, in any order. Go back to the Biotech Track to pick the one closest to your work.

intellimetrics · Arlington, VA
© 2026 intellimetrics.net
Search lessons