Software Engineer

The ten-question breakdown

Each factor is scored 0–100. Higher means easier to automate.

1. 01

   How routine and repeatable are the core tasks?

   CRUD endpoints, test scaffolding and straightforward refactoring follow repeatable patterns. Requirements discovery, architectural trade-offs and debugging unfamiliar systems remain much less predictable.

   65

2. 02

   How much of the work is purely digital (no physical presence required)?

   Most engineering inputs and outputs are digital, and repositories, build systems and deployment tools can be accessed remotely. Embedded hardware and on-site integration are exceptions.

   96

3. 03

   How available is training data for this work?

   Public repositories, documentation and programming discussions provide extensive examples. Proprietary codebases, undocumented dependencies and company-specific requirements are less accessible.

   88

4. 04

   How tolerant is the work to occasional errors?

   Draft code can be tested and revised cheaply, but production mistakes can expose customer data, interrupt operations or corrupt financial records. Tests catch only the failures their authors anticipated.

   38

5. 05

   How little regulation, licensing or legal accountability is attached?

   Most software engineering positions require no professional license. Privacy obligations, security commitments and sector-specific rules still create accountability for deployed systems.

   72

6. 06

   How little physical dexterity in unstructured environments is required?

   Mainstream software development requires essentially no physical manipulation beyond computer interaction. Robotics and hardware-focused positions constitute a more physically constrained subset.

   98

7. 07

   How weak is the requirement for human trust, empathy or presence?

   Engineers negotiate conflicting requirements, explain risks and earn permission to change critical systems. These relationships protect some roles, though many implementation tasks require little direct human contact.

   52

8. 08

   How cheap is the work to automate relative to the salary it replaces?

   AI tooling can be inexpensive relative to engineering compensation, creating strong incentives to automate bounded tasks. Integration, verification and supervising unreliable outputs substantially reduce the savings from full replacement.

   78

9. 09

   How mature is the technology already deployed in this field?

   Code completion, code generation and automated testing tools are already deployed in engineering workflows. Reliable autonomous ownership of complex systems across their full lifecycle remains immature.

   62

10. 10

    How weak are union, cultural or institutional barriers to replacement?

    Many software workplaces have limited collective bargaining protection and readily adopt productivity tools. Enterprise procurement, access controls and requirements for accountable human reviewers slow wholesale replacement.

    74