A common onboarding mistake is assuming there is one "correct" starting workspace.
There is not.
The best starting point depends on your immediate job.
If you start in the wrong place, you create avoidable rework.
Ask these in order:
1. Do I still need to compare concept directions?
If yes, start in Canvas.
2. Do I already know what needs to be built mechanically?
If yes, start in CAD.
3. Is embedded behavior the primary uncertainty?
If yes, start in Firmware.
4. Are multiple workstreams already dependent on each other?
If yes, start in System.
Start: Canvas
Why: you need direction clarity before committing to geometry or code.
Next handoff:
Canvas -> CAD for first tangible geometry.
Start: CAD
Why: your fastest path to progress is inspectable geometry.
Next handoff:
CAD -> System if complexity grows, or CAD -> Firmware if behavior integration starts.
Start: Firmware
Why: board constraints, timing, interfaces, and behavior determine design choices.
Next handoff:
Firmware -> System for module and interface coordination.
Start: System
Why: you need explicit architecture and handoff clarity before more artifact generation.
Next handoff:
System -> whichever workspace has the next execution bottleneck.
You should switch only when one of these is true:
If none are true, stay where you are and keep shipping.
1. Starting in System by default every time
System is powerful, but not mandatory for early ideation.
2. Staying in Canvas too long
Concept quality is proven by tangible outputs, not just moodboards.
3. Treating CAD as isolated from behavior
For many products, geometry and firmware choices must evolve together.
4. Ignoring System until late-stage chaos
Add structure when complexity appears, not after breakdown.
Use this quick process with your team:
1. Write current bottleneck in one sentence.
2. Pick one primary workspace for the next 48 hours.
3. Define what "done" looks like in that workspace.
4. Define trigger for switching to the next workspace.
5. Review after 2 days.
This alone removes a lot of execution drift.
No. Use System when coordination complexity justifies it.
Yes, and you should when the primary uncertainty moves.
Usually Canvas first, then CAD, then Firmware/System as needed.
If progress stalls and your current workspace is no longer reducing uncertainty, switch.
Pick your starting workspace based on the immediate job, not ideology.
Momentum comes from solving the next real bottleneck quickly and handing off with clarity.
Next step: Start your project in Haitch