From Campus to Career
“Will AI replace engineers?” is the wrong question to prepare for
The better one is how engineering roles are changing, and what a student should learn to work well alongside these tools. Practitioners describe how AI has already changed their own work — and where it quietly produces confident, wrong answers.
AI is changing the work, not only creating job titles.
Developers use it to write and review code, debug, generate tests, understand unfamiliar codebases and prototype faster. Data professionals use it to explore data, generate queries and summarise findings. Other disciplines meet it as automation, simulation, predictive maintenance, computer vision and intelligent control.
Which means students need to understand AI even when they have no intention of becoming AI specialists — and the students who intend to build AI systems need to know that using the tools and understanding how they work are entirely different skills.
What is actually changing
65 topics across 10 areas. Ten areas, and what each means for a student.
Software DevelopmentAI generates code. Somebody still has to decide whether it is any good.8
The tools help with generation, refactoring, bug-finding, tests, documentation and unfamiliar frameworks. The judgement is unchanged.
- Programming fundamentals
- Architecture
- Databases
- Security
- Performance
- Testing
- Requirements
- Trade-offs
AI can produce code. An engineer decides whether it is correct, secure, maintainable and appropriate.
Data & AnalyticsEasier to get an answer. No easier to know it is the right one.4
SQL generation, data exploration, dashboards, summaries, forecasting — all faster. The analytical questions are the same.
- Is the data reliable?
- Is the analysis meaningful?
- Are we measuring the right thing?
- Is the conclusion supported?
AI & Machine Learning CareersFor students who want to build the systems.8
Roles include AI Engineer, ML Engineer, Data Scientist, Applied AI, Generative AI, NLP, computer vision, MLOps and AI product engineering.
- Programming
- Mathematics
- Statistics
- Data
- Machine learning
- Algorithms
- Model evaluation
- Software engineering
Using AI tools and understanding how AI systems are built are different skills.
Generative AIIt creates now, rather than only classifying.5
Text, code, images, audio, video, summaries and designs — appearing in software development, support, knowledge systems, research, automation and documentation.
- Incorrect output
- Incomplete output
- Outdated output
- Bias
- Insecure output
Verification is now part of the engineering, not an optional extra.
AI-Assisted CodingThere is a difference between two sentences.6
“AI generated this code” and “I understand why this code works” are not the same claim, and an interview will find out which one is true.
- Read generated code
- Question it
- Test it
- Modify it
- Debug it
- Explain it
Used well it accelerates learning. Used badly it hides exactly what a student does not know.
AI AgentsSoftware that does things, not only answers.7
- Multi-step tasks
- Using tools
- Searching information
- Calling APIs
- Updating systems
- Analysing data
- Triggering workflows
The engineering problems here — reliability, security, cost — are where the careers will be.
Automation & Engineering JobsPredictable work changes first.7
Basic coding, documentation, repetitive testing, report generation, simple analysis and routine support. The role usually changes rather than disappears.
- Understanding problems
- Designing solutions
- Reviewing outputs
- Working with customers
- Making decisions
- Handling exceptions
- Integrating systems
What Becomes More ValuableMostly the things that were already valuable.6
When an answer is cheap to generate, the ability to evaluate it is what is scarce.
- Strong fundamentals
- Problem definition
- Systems thinking
- Communication
- Domain knowledge
- Judgement
Judgement especially: is this correct, is it safe, is it appropriate, should we use it, what follows if we do?
What Becomes Less ValuableBe careful building a career on these alone.6
- Repetitive code generation
- Copying standard solutions
- Basic content creation
- Routine data manipulation
- Manual repetitive testing
- Simple documentation
They remain useful foundations. The move is towards understanding, designing, integrating and reviewing.
Responsible UseWhat can go wrong is now part of the job.8
- Incorrect outputs
- Bias
- Privacy
- Security
- Copyright
- Sensitive data
- Over-reliance on automation
- Lack of explainability
Two students, same tools
- Student A asks AI to generate the whole project, copies it, gets it working, and cannot explain it
- Student B uses AI to understand concepts, generate examples, debug and compare approaches — and still builds it
- Both have a working project. Only one survives an interview
- AI can help you learn, or hide what you have not learned. The difference is whether you build
How to actually use it
- As a learning partner: explain concepts, give examples, create practice problems, check understanding
- As a coding assistant: review code, find bugs, explain errors, suggest tests, improve readability — then verify
- For projects: chat interfaces, recommendation systems, document search, computer vision, automation, agents
- Move from “I know how to use ChatGPT” to “I understand how AI becomes part of a useful system”
What building with AI actually involves
- APIs
- Prompting
- Model selection
- Retrieval
- Databases
- Tool integration
- Agents
- Evaluation
- Guardrails
- User experience
- Cost
- Reliability
AI in the hiring process itself
- AI-assisted screening
- Automated assessments
- Online coding platforms
- Questions about AI-assisted development
- And the question underneath: can you solve problems without depending on it?
- Students should be able to say how they use AI and where they verify its work
How the session runs
Practitioners describe how AI has already changed their own work, role by role — software, data, cloud, security, testing, product and core engineering. One demonstrates it live during real engineering work, including where it produces a confident and wrong answer. Students then ask the career questions directly, and each leaves naming one AI capability to learn, one fundamental to strengthen, one project to build and one way to use it more responsibly.
What your students leave with
- A clear view of how AI is changing engineering roles, not just adding new ones
- Where AI is genuinely being used in industry today
- Which fundamentals become more valuable rather than less
- How to use AI as a learning partner without outsourcing understanding
- How to build AI into a project rather than demonstrating an API call
- Where responsible use actually matters
- One capability to learn, one fundamental to strengthen, one project to build
Scheduled sessions
Nothing scheduled yet
Sessions are arranged with a college once a date is agreed. Ask us and we will find the right person for it.
A student rather than a college? See what is coming up, or ask your placement team to host this.
Not by ignoring AI, and not by depending on it for everything.
Tell us who your students are and what stage they are at. Sessions are free for participants.