Applied AI-Powered Bioinformatics Tools

Usage & Billing Summary Dashboard

Last reviewed

Beginner

What you'll learn

~40 min
  • Ask an AI tool, in plan mode, for a small program that turns a month's usage log (the booking system's record of who used what, when and on which account) into a summary of each lab's charges
  • Give it your core's rules for what is charged, at which rate and for what time, and read the plan for lines it holds for you instead of charging: entries logged twice, no account, an account that has ended, a rate that does not match
  • Check the summary's totals and held lines against figures worked out for the practice files
  • See why totals that balance cannot show who used an instrument, and who can

The story, and what you will end up with

It is Wednesday 4 November 2026. You keep the usage records for a core facility that runs a confocal microscope, a flow analyser, a cell sorter run by its own staff, and a sample preparation service, and at the start of each month you prepare the charges for the month before. October’s bookings are closed in the booking system, and users had until yesterday to correct their own bookings; none was changed after that. You have already done the checks that come before any summary: the export’s 27 entries match the booking system’s own October report, the rate schedule is the approved one in force since 1 July, the list of accounts and purchase orders is current, and nothing was left over from September.

What comes next is the part that goes wrong at month end: a session entered twice, a booking at the wrong rate, a session with no account or on one that has ended, a session that ran past midnight, a booking that was cancelled but still has its times. Until now you found these by reading a spreadsheet line by line.

By the end of this lesson you will have a small program on your own computer, the charge summary, that reads October’s three files and makes a page you can print: each lab’s charges by account or purchase order, with every line that has a question on it set apart for you to settle, and what each would cost. It is a draft for checking. It sends no invoice and posts nothing to any financial system; before anything is billed, you and the person who approves the month’s charges in your facility, such as the facility manager, check it against the usage log and the approved rates. You build it by asking your AI tool in plain English, and you check it against figures this lesson gives you.

🔍Words used in this lessonclick to expand
  • Usage log: the booking system’s record of who used which instrument or service, when, for how long or how many samples, and on which account.
  • Rate category: the price band a charge falls in. The practice files use three: Internal, External academic (another university) and Industry (a company).
  • Account: what an internal charge is paid from, here written like ACCT-3101. Users from outside are billed on a purchase order (PO) instead. Each has dates it is valid between.
  • Held: set apart for a person to settle, and in no total, until they have.
  • Draft: what the summary is. It is checked before any charge goes out, and changed if a check finds something.

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/08-usage-summary && cd ~/bio-workbench/08-usage-summary

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/08/usage-october.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), three files are in the data folder (usage-2026-10.csv, accounts.csv and rates-fy2027.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 usage-october.zip in your browser, check where it went (the browser’s list of downloads shows the folder), and tell your AI tool, for example: “usage-october.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 usage-october.zipclick to expand

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

usage-2026-10.csv, the booking system’s export for October: 27 entries, U-2601 to U-2627. Each has a date, the resource (an instrument or a service), the user, their lab, the account or purchase order it was booked to, a start and end time (instruments) or a number of samples (sample preparation), a status (Completed or Cancelled), and the rate category the booking carries.

accounts.csv, what each lab may charge: five internal accounts for three labs here, and two purchase orders for the users from outside. Each has a rate category and the dates it is valid from and to.

Account or orderLabRate categoryValid
ACCT-3101Arendt labInternal1 July 2025 to 30 June 2027
ACCT-3102Arendt labInternal1 October 2024 to 30 September 2026
ACCT-3205Baptiste labInternal1 January 2026 to 31 December 2028
ACCT-3206Baptiste labInternal1 April 2025 to 31 March 2027
ACCT-3308Costa labInternal1 September 2025 to 31 August 2027
PO-77120Ferreira lab (another university)External academic1 July 2026 to 30 June 2027
PO-88415Quillmark Bio (company)Industry1 September 2026 to 31 August 2027

rates-fy2027.csv, the practice core’s rate schedule, in force from 1 July 2026:

ResourceCharged byInternalExternal academicIndustry
CONF-1, confocal microscopehour$40$60$100
FLOW-A, flow analyserhour$32$48$80
SORT-1, cell sorter, run by core staffhour$88$132$220
PREP, sample preparationsample$12$18$30

This practice core’s rules: the rate follows the account or order that pays, at its rate category; an instrument is charged for the booked time from start to end, rounded up to the next quarter-hour; a cancelled booking is charged nothing. Your core’s rules are in its own procedures, and how rates are set and approved is decided by your institution, not by whoever prepares the summary.

The entries you will meet further down:

EntryWhat it is
U-2604, U-2623A. Ito’s two sessions on ACCT-3102, which ended on 30 September
U-2610A confocal booking, 9:00 to 12:00 on 12 October, cancelled
U-2611, U-2612The same flow analyser session, S. Brennan, 13 October, 10:00 to 11:15, entered twice
U-2616A flow analyser session with no account
U-2619A confocal time-lapse from 22:30 on 21 October to 01:00 the next morning
U-2620A cell sorter session on Quillmark Bio’s purchase order, booked at the Internal rate
U-2627A confocal session from 9:00 to 10:50
U-2601An ordinary confocal session, J. Moreau, Arendt lab, 1 October, 9:00 to 11:30

Everything in these files is made up for this course: every person, lab, company, account, purchase order and rate, and this practice core’s rules above. 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 keep the usage records for our core facility. From the October files in data, build a summary of each lab's charges that I can check before anything is billed. Ask me what you need first.

Your AI tool usually reads all three files and comes back with questions. It may also point out some of the entries this lesson is about before you ask.

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
U-2604 and U-2623 are on an account that ended on 30 September. Hold them, or move them to the lab’s other account?Hold both for me. I’ll ask the lab which account to use.
Do you charge for cancelled bookings?Charge nothing for cancelled bookings, and list them.
U-2611 and U-2612 look identical. Charge once?Hold them as one question. I’ll check the instrument’s own log.
U-2616 has no account. Hold it, or do you know which?Hold it for me.
U-2619 runs from 22:30 to 01:00. Is that 2.5 hours?Yes, U-2619 is 2.5 hours. It ran past midnight.
U-2620 is booked at the Internal rate on an Industry purchase order. Which rate decides?The rate follows the account that pays, but hold U-2620 for me, showing both amounts.
How are part hours charged: exact minutes, rounded, a minimum?Round each session up to the next quarter-hour.
What currency?US dollars.
A page, a spreadsheet, or both?A page I can open in my browser and print.

If it asks several questions at once, answer the ones about what is charged, at which rate and for what time from your core’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.

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. Each charge is worked out by your core’s rules: the booked time from start to end, rounded up to the next quarter-hour, including a session that runs past midnight, or the number of samples; times the rate for the rate category of the account or order that pays.
  2. Anything the program cannot settle is held for you, with what it would cost, and is in no total: an entry that appears twice (one question, not two charges), no account, an account used outside its dates, and a booking whose rate category differs from its account’s.
  3. Cancelled bookings are listed and charged nothing.
  4. Totals are by lab, and within each lab by account or purchase order, added from the line charges, with the month and the rate schedule’s date printed on the page.
  5. It is a program you can run again next month, and it finds the lines to hold from the files themselves, not from a list of this month’s entries.
  6. Nothing is sent or posted, and your three files are read and left as they are.

A first plan often leaves one of these out, most often the month and the rate schedule’s date in item 4, or item 5. Say so in a sentence, in your own words, for example:

Also print the month and the rate schedule's date, and make it a program I can run next month that finds the lines to hold itself.

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 run what it built and tell you how to run it yourself; either way, the next step asks it to. 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, ask it what that is for, and to remove it if you do not want it there.

Look at what it built

Ask it to run what it built:

Run it and show me the summary.

It may show the figures in the terminal and say where the page is saved, or start the page and give you an address to open. To see the page itself, ask: “Open the summary in my browser.” If it gives you a file location or an address without opening it, copy that 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 summary page if it is still running.”

Your summary will be worded and laid out in its own way. Check it for these:

  • It is for October 2026, and it names the rate schedule it used, in force from 1 July 2026.
  • All 27 entries are accounted for: 20 charged, 6 held, 1 cancelled.
  • The charged total is $2,288.00. By lab: Arendt lab $384.00, all on ACCT-3101; Baptiste lab $500.00, $260.00 on ACCT-3205 and $240.00 on ACCT-3206; Costa lab $560.00 on ACCT-3308; Ferreira lab $294.00 on PO-77120; Quillmark Bio $550.00 on PO-88415.
  • Held, with what each would cost, and in no total: U-2604 ($80.00) and U-2623 ($48.00), on ACCT-3102 after it ended; U-2611 and U-2612 as one question, $40.00 for the one session; U-2616 ($48.00 at the Internal rate it was booked at), no account; U-2620, $440.00 at the purchase order’s Industry rate and $176.00 at the Internal rate it was booked at. If U-2620 sits in Quillmark Bio’s total, making it $990.00, it has been charged, not held; tell your AI tool to hold it.
  • U-2610 is listed as cancelled and charged $0.00.
  • U-2619 is 2.5 hours, $100.00. If it shows a negative time or a charge below zero, the session past midnight has been worked out on clock times alone.
  • U-2627, 9:00 to 10:50, is charged as 2 hours, $80.00, rounded up as you said. Charged by the exact minute it would be $73.33, and the Costa lab $553.33; that is another rule a core may use, and it is why the page says which rule it used.
  • When you ask “Does the program name any entry, account or lab from these practice files? If it does, make it work those out from the files instead.”, it says it does not, or changes the program so it finds them from the files and tells you what it changed. A program built in a conversation may keep your answers about this month’s entries written into it, and next month it would hold the wrong lines.
  • When you ask “Have my usage files changed since the checkpoint?”, it compares them with the checkpoint and says no.

Notice what the totals cannot show

The held lines are the ones the files themselves disagree about. Now try a mistake the files would not disagree about: a session entered under the wrong person.

Copy the usage log into this folder. In the copy only, change U-2601 to S. Brennan, Baptiste lab, account ACCT-3205. Show each lab's total and the grand total from the copy beside the original, and anything held.

Before you read on: the copy now charges the Baptiste lab for a session the Arendt lab ran. What do you expect the summary to say?

Nothing that would catch your eye. The grand total is $2,288.00 in both. The Arendt lab goes from $384.00 to $284.00 and the Baptiste lab from $500.00 to $600.00: U-2601 is 2.5 hours on the confocal at $40, $100.00, now on the other lab. The held lines are the same as before, and nothing new is held, because the changed entry agrees with every other file: S. Brennan is in the Baptiste lab, and ACCT-3205 is a Baptiste account in force on 1 October.

A summary can check its files against each other. It cannot know who was at the microscope that morning. The instrument’s own login record, the staff member on duty, or the lab itself, reading its own lines, can.

That is why the summary stays a draft until every held line is settled, and why each lab sees its own lines before anything is billed, if your core’s procedures provide for it.

Totals that balance show that the sums agree, not who used the instrument: only someone who knows who was there, or a record kept apart from the bookings, can show whose charge it is.

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

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 summary against the list above, not against this.

I made the copy and changed U-2601 in it, which moves $100.00 from
Arendt lab to Baptiste lab. The grand total stays at $2,288.00 and
the same six entries are still held.
U-2601 is 2.5 hours on the confocal at the Internal rate, which is
$100.00. Both accounts are Internal and both were valid on 1 Oct, so
the amount doesn't change; it just moves from one lab to the other.

The labs, people, accounts and rates it talks about are made up for this course.

How real billing differs

The practice files are small and their rules are made up. A real month has more in it:

  • Your core’s rules are its own. Check what yours say: whether to charge the booked time, the time the instrument was logged in to, or the greater of the two; whether there is a minimum charge or rounding to the half-hour; whether late cancellations and no-shows are charged, which then need a status of their own. The summary follows your core’s written rules, and the page says which ones it used.
  • More kinds of line. Check which of these your log carries: staff-assisted and unassisted rates, training and courtesy hours, subsidies, after-hours rates, credits for a session cut short when an instrument failed, and staff maintenance blocks in the same log as user bookings.
  • More kinds of account trouble. Check which of these your procedures cover: an account split between two sources, a user who belongs to two labs, a booking on an account that belongs to another lab, an account that is open but not approved for this kind of charge, a purchase order with less left on it than the month’s charges, two bookings on one instrument at the same time. Have the program hold a service line with no sample count, not charge it as zero.
  • Time is not always clock time. A session that runs overnight where the clocks change, the night they change, is an hour longer or shorter than its start and end times say. A session that runs past midnight on the last day of the month needs your core’s rule for which month it belongs to.
  • Held lines need an owner. Your procedures say who settles each one, and whether it is settled before the month is closed or carried to the next month with a written reason. On this summary a held line is in no total, so it never goes to a lab as a charge or an estimate.
  • After the charges post, compare what landed in the financial system with the summary you approved.

Keep going

Small changes do not need plan mode. Say each in a sentence, then run it again:

  1. “Also hold any line whose lab is not the lab its account belongs to.” None of October’s practice entries is booked to another lab’s account, so nothing new should be held; in a real month this catches a session charged to the wrong lab’s account.
  2. “Make a separate page for each lab with only its own charged lines, for its PI or lab manager to confirm.” Held lines stay off it; you take those up with the lab yourself. Whether and when labs see their lines before charges post is your core’s procedure; this gives you something to send if it does.
  3. “Show each instrument’s charged hours for the month beside its total.”
  4. “Save a checkpoint.” Your AI tool saves a snapshot you can return to.

Your own month

When you want a summary of a real month, use the finished summary program. It runs on your real files by itself, with your AI tool closed, so your AI tool never sees them. A real usage log names people and labs, and its charges are your facility’s business. 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 files: its columns will have other names, and it will carry real people’s names and real accounts. Your accounts list and rate schedule may not come as files at all; you type them into blank sheets. Do these steps in order; the first six 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. A usage log names people and labs and carries account numbers; it is your facility’s to decide on, with whoever your institution says looks after its financial records. Go on only if they allow this use.

  2. Decide on a folder of its own for your real files, outside this course folder, for example ~/core-billing. 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 core-billing in my home folder.”

  3. Ask your AI tool for blank sheets to fill in, for example:

    Make a blank rate schedule and a blank accounts list my spreadsheet program can open, both in ~/core-billing. Show the summary working on one made-up row in each, then clear the rows.

    If you chose another folder, put its name in place of ~/core-billing, starting with ~/ as here (~ stands for your home folder). Look for where it says it saved the two files, and ask what any heading means if you are not sure. Check that it says the made-up rows gave a charge and that both sheets are back to their headings only. If it offers to copy the practice accounts or rates into them, say no.

  4. Open your booking system’s export in your spreadsheet program and note its column names and how it writes dates, times and statuses. Tell your AI tool those, and give it two rows you make up in the same shape, with made-up names, accounts and times, never a real row. For example: “My booking export has the columns Booking ID, Instrument, User, Group, Account, Start, End, Quantity and Status, with dates like 10/13/2026 2:00 PM. Here are two made-up rows in the same shape. Change the summary to read files like that.” Then paste the two rows you made up, and say “Save a checkpoint.”

  5. The summary still follows the practice core’s rules: quarter-hours rounded up, cancellations free. Tell your AI tool each charging rule from your core’s written procedures, one sentence at a time, for example: “Our core rounds each session up to the next half-hour, with a one-hour minimum.” Then ask: “Show me the charge for one made-up session under each rule I gave you.” Work each one out by hand, and say “Save a checkpoint.”

  6. Ask your AI tool for the exact line, for example: “Write me the exact line to run this myself on the files in ~/core-billing, saving the summary there.” 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
    cd /home/dana/core-billing && python3 /home/dana/bio-workbench/08-usage-summary/scripts/build_charges.py usage.csv accounts.csv rate-schedule.csv charges-summary.html

    Read it left to right. cd /home/dana/core-billing goes into your billing folder, and && means “then”. python3 runs a Python program: next comes the summary program, in this lesson’s folder; then the month’s usage export, the accounts list and the rate schedule, all in core-billing; last, the summary page it writes there. /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/. Each month you change the file names to that month’s. The line runs the program where it is, so leave this lesson’s folder in place. If you are unsure about any part, ask your AI tool now, while it is still open.

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

  8. Fill in the blank sheets from your approved rate schedule and your current list of accounts, save your month’s export into that folder with the name the line uses, paste the line from your note at the $ or % prompt and press Enter. The summary appears beside your real files, not in the course folder. 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.

  9. Before you rely on it, check every rate you typed against the approved schedule, the entry count and grand total against the booking system’s own report for the month, a few charged lines by hand, including one under each of your own rules, and every held line against the records. A matching total is not enough by itself, for the reason this lesson showed.

Try another case

Before the quiz, write two or three sentences for yourself. In November’s summary, the Costa lab’s total is $40 higher than its PI expected, and the lab manager says one flow analyser hour on 9 November was not theirs. Nothing on the summary is held. What do you check, and what do you tell the lab?

A good answer does not change the line because the lab asked, and does not refuse because the summary raised no flag. It looks for a record kept apart from the bookings, such as the instrument’s own login record or the staff member on duty, to find out who used the analyser that hour. It tells the lab the line is held while that is checked, and it changes the charge, or explains it, from what that record shows.

Key takeaways

  • A short ask is enough. Say what you keep, which files, and what you want; its questions fill in your core’s rules.
  • Your core’s rules come from your procedures, never from the AI tool’s suggestions: what is charged, at which rate, for what time.
  • Every question goes to a person. Entries logged twice, no account, an ended account, a rate that does not match: held, with what each would cost, in no total.
  • Totals that balance show that the sums agree, not who used the instrument.
  • Practice data stays practice data: real usage logs go to the finished program, with your AI tool closed, after the data check.
KNOWLEDGE CHECK

Your summary's grand total matches the booking system's October report to the cent, and nothing on it is held. A lab manager says one confocal session charged to their lab was not theirs. What does the matching total tell you about that session?

What’s next

The next lesson is about planning a mouse colony’s breeding for a study: what a line’s breeding pairs can be expected to give, and by when.

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