How to start a career in AI (without a CS degree)

A computer science degree is not required for every AI role. Entry requirements depend on the employer, location and type of work; research and some engineering roles have substantial technical or academic prerequisites. Start with actual vacancies and build evidence of the skills they require. A course certificate can record learning, but it does not guarantee an interview or a job.

First, choose a lane

AI work includes application engineering, machine learning, research, product management and adoption or consulting. Application engineers build on existing models and need programming, debugging, testing and data skills. Product and consulting paths emphasise domain expertise and decision-making, with technical literacy appropriate to the service. Choose a path that fits your experience and verify its requirements before buying training.

Skills to practise for application engineering interviews

• Retrieval-augmented generation from end to end: chunking, embeddings, retrieval, and the reasons a naive implementation breaks in production.
• When to fine-tune a model versus when a better prompt or better retrieval solves the problem more cheaply.
• Agent design, with particular attention to where an agent fails silently.
• Evaluation: how a candidate would know a system got worse, not merely whether it feels better.
• Cost and latency trade-offs: compare the cost of the complete workflow with the value it delivers, and show how quality changes when you optimise it.

Why tutorials do not lead to job offers

A tutorial chatbot demonstrates the ability to follow instructions. It does not demonstrate the ability to debug the case where a retriever confidently returns the wrong document, nor does it demonstrate an understanding of why that happens. Interviewers at companies doing serious AI work probe for exactly that judgment. It is the actual job.

The dangerous failure in an AI system is the one that looks like it worked.

Build a portfolio that demonstrates failure analysis

A focused project with reproducible evidence is more useful than several unexplained demos. Start with fifteen to twenty representative cases to find initial failures, including missing evidence and ambiguous requests. Keep fresh cases for the final evaluation, run repeated comparisons and report counts and limits. A small portfolio test set does not establish production reliability.

A twelve-week practice plan, adjusted to your starting point

If you already program comfortably, use weeks 1–3 to build a small document assistant and an initial evaluation set. In weeks 4–6, improve retrieval and source checking; add tools only where the task needs them. In weeks 7–9, test fresh cases, record cost and latency, and document failures. In weeks 10–12, refine the portfolio and apply to suitable roles. If programming is new, begin with APIs, JSON, Git, debugging and tests; allow additional time. This is a practice schedule, not a promise of employment within ninety days.

Where to learn each piece

The free Course Parse course covers the fundamentals of using AI tools well. It is worth completing first if the basics are not yet familiar.

For the technical path, AI Engineering covers retrieval, fine-tuning, agents, evaluation, operating costs and job-search preparation. Its four capstones are a web application, a mobile application, a verified data analysis, and an agent with a written evaluation. The certificate requires completion of the lessons and approval of all four projects. AI Product Management, AI Consulting and Prompt Engineering offer different routes; choose according to the work you want to do.