No one to convince
No IT ticket, no committee, no pilot program. If it works on Tuesday, it’s part of your workflow on Wednesday.
A membership for solo structural consultants
You’re the whole firm. Your calcs should pull their weight.
Structural Python is a working community for structural engineers who run their own practice — or think like they do. Learn to build calculation workflows that link inputs to outputs, chain one design into the next, and produce the finished calc package. All on open‑source tools we build ourselves.
- Since 2020 teaching structural engineers Python
- ★ 5.8k on GitHub for handcalcs
ProjectLaneway house
SheetB‑3
ByYou
CheckedAlso you
Floor beam B‑3 · grid C
L = 6.10 m
wD = 4.20 kN/m
wL = 6.00 kN/m
wf = 1.25 · wD + 1.5 · wL
= 1.25 · 4.20 kN/m + 1.5 · 6.00 kN/m = 14.25 kN/m
Mf = wf · L28
= 14.25 kN/m · (6.10 m)28 = 66.28 kN·m
Vf = wf · L2
= 14.25 kN/m · 6.10 m2 = 43.46 kN
Who this is for
Automation pays best when you own the practice.
Large firms talk about automation for years. A solo consultant can have it working by next week. That’s why this community is built around one kind of engineer: the one who signs the drawings, sends the invoice, and keeps what they save.
Every hour saved is yours
On a fixed‑fee job, a calc that takes twenty minutes instead of three hours is pure margin. You don’t share the upside with anyone.
It’s a business expense
The membership is a tool for your practice, the same as your analysis software. Run it past your accountant — but you knew that.
Not a solo consultant? If you’re the engineer in your firm who just quietly gets on with it, you think like one — and you belong here too.
Why now
LLMs can write the code. They can’t design your workflow.
Writing Python is no longer the hard part. Ask any LLM for a beam function and you’ll get one. What it won’t hand you is the structure: how a job’s inputs flow to its outputs, how one calculation feeds the next, and how all of it lands in a calc package you’re willing to stamp. That’s what turns scattered scripts into a practice that gets faster every month — and it’s what we work on together.
Scripts you prompt into existence
- Solve today’s problem
- Live in a folder called
final_v3 - Spit out a number you retype into Word
- Break when the next job is a little different
A framework you build on
- Solves the whole class of problem
- Lives in a job template you reuse
- Outputs feed the next calc and the report automatically
- Bends to small, custom, one‑off jobs — not just the repetitive ones
The method
Chaining: outputs become inputs.
Most automation talk is about big, repetitive jobs. But most of a small practice’s work is small, custom and one‑off. More scripts won’t fix that. A data framework will.
The ASET template is a job‑execution structure that links every input to every output, so the results of one calculation become the inputs of the next. ymprint orchestrates the job. handcalcs renders the calculations. The calc package builds itself.
Change a load at the top, re‑run, and every downstream member — and every page of the report — updates with it.
-
Inputs
Loads & geometry
job.yml -
Calc
Beam B‑3
→ Vf, Mf -
Calc
Column C‑2
Pf ← B‑3 -
Calc
Footing F‑1
Pf ← C‑2 -
Output
Calc package
report.pdf
Open source
We build the tools. You build with them.
Structural Python doesn’t just repackage other people’s software. We do the original research and write the open‑source libraries the method is built on — foundations for the big repetitive jobs and the small one‑offs. They’re free for everyone. Members learn to use them properly.
handcalcs
v2.0Write a calculation in Python and get it rendered as if you wrote it by hand. Version 2.0 goes further: a calculation rendering framework that works directly in document production.
ymprint
NewDesktop publishing with YAML. Write report content as data, render to PDF in an instant, and orchestrate the whole job through the ASET template.
forallpeople
An SI units library that’s your daily driver for calculations. Units carry through every step, so kN·m stays kN·m.
papermodels
A prototype structural data model for buildings: one source of truth that your calculations can read from and write to.
loadtracer
Traces loads through a building structure, so the results flow straight into the member calcs downstream.
Browse it all
Every library, starter kit and experiment.
github.com/StructuralPython → github.com/connorferster →
Who builds this
Engineer first. Toolmaker second. Teacher since 2020.
I’m Connor Ferster, a structural engineer in Vancouver, BC. I wrote handcalcs because I was tired of retyping my own calculations, and it grew into one of the most widely used Python tools in structural engineering.
Since 2020 I’ve taught structural engineers to automate their work. Over that time I’ve learned who gets the most out of it: the engineer who can put a new tool to work the same afternoon and keep every hour it saves. This community is built for you.
- 2020
- teaching structural engineers Python
- 5.8k
- GitHub stars on handcalcs
- 5
- open‑source building blocks
- 2×
- group check‑in calls a month
Membership
Priced like a tool, not a degree.
What’s $97 to your practice?
Put in your own numbers.
$
/ hr
39 min
of your billable time pays for a month
$600
of time recovered every month — 6.2× what it costs
One job where the calc package takes an afternoon instead of three days can cover a year of membership.
Structural Python Membership
$97/ month
Less than one billable hour.
- The community spaces — ask, share code, get unstuck
- Short group check‑in calls, twice a month
- The full library of past Structural Python training
- New training as it’s released: handcalcs v2.0, ymprint, the ASET template and chaining
Monthly subscription. Cancel anytime.
Questions
Before you join
Do I need to know Python already?
No, but you need to be willing to learn. The full library of past training is included, so you can start from the beginning and the community is there when you get stuck.
Can’t I just ask an LLM?
Use one! Members do. An LLM is a great way to write a function. It won’t give you a way to organize a job so inputs, calculations and the report stay linked. That framework is what we teach, and it makes your LLM‑written code far more useful.
Is this only for big, repetitive jobs?
No. That’s the point of chaining and the ASET template. Once a job is structured as linked inputs and outputs, even a small custom one‑off goes faster, because you reuse the pieces rather than the whole job.
I work in a firm. Is this for me?
If you have the freedom to change how you do your own work, yes. The community is built around solo consultants because they can act right away, but anyone who thinks that way will fit in.
What happens on the check‑in calls?
They’re short group calls, twice a month. Bring what you’re building or where you’re stuck, and see what other members are working on.
Can I cancel?
Yes. It’s a monthly subscription and you can cancel anytime.
Make your next job the first one that builds itself.
Join for $97/monthMonthly subscription. Cancel anytime.