Sessions need a laptop or desktop.
If you’re on your phone now, that’s fine. Enrol now and we’ll email you a link so you can pick up when you’re at a computer.
Python · Foundations
Answer a manager’s questions in a notebook you can rerun.
What you’ll end up with
One finished analysis notebook. The cells you wrote, runnable first to last, answering what was sold, to whom, when, where — and how much of the export can be trusted.
10½
hours in all — about 7½ guided, 3 on your own
6
sessions. Each designed to be completed in one sitting
Free
No payment and no card required.
Finishing gives you a completion record showing the work you built, and all three levels earn the Alongrail Certificate in Python. See what you’ll build →
You don’t need prior experience with Python in JupyterLite. There’s no payment and no card required.
From two raw files to a finished notebook
01105 min
Open the file and find out what you have
You have Python running in your browser with nothing installed, both exports read in, and the size and shape of each one written down before anything is claimed about them.
0295 min
Ask for the rows and columns you actually want
You can pull out the columns you care about, keep only the rows that meet two conditions at once, count what you kept, and put a table in an order you chose rather than the one it arrived in.
03140 min
Find out what the values really are
You have established what the reader actually made of this export — which money is text, which dates it refused, how many spellings one category really has, and the two different ways a value can be missing — and you have fixed each one in a column of its own.
0490 min
Count and total it, and say how precisely
You can total, average and find the largest of a column, say to what precision, and say which rows each figure covered and which it quietly skipped.
05108 min
One row per group, and the duplicates nobody mentioned
You can produce a figure for every category at once, name those figures, order them, and find the repeated rows that were sitting inside every total you produced last session.
0697 min
Two tables at once, and what the join lost
You can bring the customer file alongside the order file, say exactly how many orders the join dropped and why, fix the cause rather than accept the smaller answer, and produce the figure per region the sales lead asked for.
What you need before you start
A laptop or desktop
Sessions need a laptop or desktop. If you’re on your phone now, that’s fine. Enrol now and we’ll email you a link so you can pick up when you’re at a computer. Python in JupyterLite runs in the browser, on Windows or a Mac.
Chrome or Edge
Alongrail runs in the browser too, in its own tab beside Python in JupyterLite.
Python in JupyterLite
Python in JupyterLite runs in your browser. Nothing is installed.
The course files
Everything you work with is in the download.
What’s free
The whole course: every session, both parts, the files, and your record. Alongrail charges nothing for it. No payment, and no card at any point.
The software
Nothing is paid to anyone at any point: no licence and no account.
The data you’ll work with
Every file is invented — the companies, the people and everything in them are made up for the course, and the files are deliberately messy, the way real business data is. Cleaning them is part of the work.
Before you start
The course covers Python in JupyterLite, in your browser.
- A current desktop browser. Nothing is installed, no account is created and nothing is signed in at any point.
- The address of the notebook page, jupyter.org/try-jupyter, which you open as an ordinary web page
- orders.csv and customers.csv, downloaded from the course page into one folder you can find again
- The five start notebooks for the guided run, downloaded from the same place into the same folder
- The second company's two exports and five start notebooks, downloaded to the same folder. Their names carry the company, so both sets can sit in the page together without one replacing the other
Alongrail teaches the software and gives you a record of what you built, with a link to the work itself. What happens with it afterwards is yours. We don’t place people and we take no part in anyone’s recruitment process.
Enrol free and start now.
Sign in, confirm the one thing the course needs, and session 1 opens. No payment, and no card at any point.
If something goes wrong on the way, press “Something’s wrong” in the panel, or write to hello@alongrail.com — we reply within one working day.
Frequently asked questions
What you build
Open a file somebody else exported, in a browser, with nothing installed and no account — and answer the questions a manager actually asks of it, in code another person can read and run again next month, checking each answer before you send it.
An analysis notebook you keep: the cells you wrote, still runnable from the first to the last, answering one question per session about a wholesale supplier's two years of orders — what was sold, to whom, when, where, and how much of what the export says can be trusted.
How a session works
- Open Python in JupyterLite
- Share your screen
- One instruction
- You do it
- It confirms it
- Next
You open Python in JupyterLite and share your screen. Alongrail reads what’s there and gives you one instruction. You do it. It confirms the step landed, then gives you the next one. If something doesn’t match what you see, one button gets you back on track.
Guided
In the Fields list, tick Region. The chart fills in.
One field on the axis, one measure as the value.
then
On your owna different company’s data
Same job, a different company's files. Alongrail stays quiet until you ask.
Guided, then on your own.
The second part of every session uses a different company’s data, with the same kind of mess in different places. Alongrail stays quiet until you ask. It’s the work your record points to.
About 10½ hours in all·about 7½ guided·about 3 on your own
What you earn — the record, the certificate and a sample
Completion record
- Name
- Your name
- Course
- Python Foundations — answer a manager's questions from a file you did not make
- Completed
- The date you finish
- Project
- The notebook you built on your own →
- Skills demonstrated
- Reading a file somebody else exported into Python and establishing its size and shape before claiming anything about it · Selecting columns and filtering rows on several conditions at once, in an order chosen on purpose · Establishing what values really are — money read as text, two kinds of absence, one category written many ways, and dates the reader refuses — before computing with them · Counting and totalling, and knowing which count skips the rows with nothing in them · Answering a question per group, and finding repeated records that were inside an earlier total · Joining two tables and stating what the join lost · Checking an answer against the row count before reporting it
What was built
An analysis notebook you keep: the cells you wrote, still runnable from the first to the last, answering one question per session about a wholesale supplier's two years of orders — what was sold, to whom, when, where, and how much of what the export says can be trusted.
The 6 sessions, and when each was completed on their own
Each session was completed twice: once with step-by-step guidance on one company’s data, then on a second company’s data from a brief, with guidance only when asked. Every check passed on both. Alongrail records the sessions, the checks and the dates; the dashboard itself is the learner’s own, hosted where they chose.
A completion record for Python Foundations, now. It names the course, the date you finished, the skills you demonstrated, and links the notebook you built on your own. It’s private until you publish it, and you can unpublish it at any time.
The record is a description of work you did, not a claim about you. Anyone who opens it can see the work for themselves.
The Alongrail Certificate in Python when all three levels are complete. Foundations is the first of three. Practitioner and Professional are in development, with no dates. See the track →
Your screen stays yours
It looks when you ask. Never otherwise. Images of your screen are not stored. What Alongrail sees →
Nothing you build is public until you publish it. Your record is private until you say so, and you can unpublish it at any time.
What’s Python in JupyterLite?
Python in JupyterLite is the software this course is built in. It runs in your browser.
You won’t learn it by watching hours of tutorials. You’ll learn it by building something real with it.
Who you’re building it for
Kestrel Provisions Ltd supplies offices across four regions with packaging, cleaning materials, catering, stationery and workwear. Someone has exported two files out of the order system and sent them to you: one row per order, and one row per customer. The sales lead wants to know how the two years went, which categories carry the business, which customers are worth a call and which region is worth the next hire. Nobody has checked the export and nobody will. You have no software budget, no access to the order system and nobody to ask. What you have is the two files and a browser.
Your job is to build the notebook that answers it.
Do I need to know Python in JupyterLite already?
No. The course assumes you’ve never opened it.
What if I get stuck?
Press the button in the panel and say what you see. Alongrail works out where you actually are and picks the instruction back up.
Why 10½ hours, when other courses are 40?
Because the time is spent doing the work rather than watching it. About 7½ hours is guided; the rest is you building the same thing again on your own.
What does Alongrail see on my screen?
Only the window you choose to share, and only when you ask it to look. Images of your screen are not stored by Alongrail. What Alongrail sees →
Can I use a Mac?
Yes. Python in JupyterLite runs in the browser.
What if I miss a week?
Nothing closes. The dates are there to help you keep moving, not to catch you out.
What’s the “on your own” part?
The second half of every session: the same work on a different company’s data, with Alongrail quiet unless you ask. Both parts are needed for your completion record.
Is it really free?
Yes. No payment, and no card at any point.
Who can see what I build?
Nobody, until you say so. Your record is private until you publish it, and you can unpublish it at any time. What you build stays in your own files or account; where a step publishes something to the software’s own site, the step says so and says who can see it.
Where this course sits
Python Foundations is Foundations in Python. The track has three levels; each one you complete earns its own record, and all of them together earn the Alongrail Certificate in Python. All the courses →