Careers in and Around AI

When people think about AI careers, they typically picture one role: the machine learning engineer or researcher who writes the algorithms. That role is real and important — but it represents a small fraction of the work that shapes how AI affects the world. The landscape of AI-related careers is far wider, and many of the most consequential positions require no deep mathematical background at all.

This matters for two reasons. First, it means that students with a wide range of interests and strengths have genuine paths into meaningful AI work. Second, it means that the question of who fills these roles matters enormously for what AI becomes. A profession dominated by people who share similar backgrounds, values, and experiences tends to build systems that reflect those backgrounds and miss what lies outside them.

The Technical Core

Machine Learning Research Scientist: Designs new algorithms, architectures, and training methods. Advances the fundamental capability of AI systems. Typically requires a PhD or strong graduate-level background in mathematics, statistics, and computer science. Publishes in venues like NeurIPS, ICML, and ICLR. ML Engineer: Implements, trains, optimizes, and deploys ML models at production scale. Bridges research and application. Requires strong software engineering alongside ML knowledge. Writes code that other engineers' code depends on. Data Engineer: Designs and maintains the data pipelines, storage systems, and processing infrastructure that feed ML systems. Often invisible, but without high-quality data infrastructure, no model gets trained. Requires strong systems and software engineering skills. Computer Vision and NLP Specialists: Domain-specific engineers and researchers focused on perception (images, video, 3D) or language (translation, generation, understanding). Often combine ML foundations with deep domain expertise. AI Safety Researcher: Studies how to ensure AI systems behave reliably, honestly, and in alignment with human intentions even as they become more capable. Combines ML with philosophy, formal verification, and interpretability research. A rapidly growing and critically important field.

The Broader Ecosystem

AI Ethics Researcher and Practitioner: Studies the ethical implications of AI systems and advises organizations on how to design, deploy, and evaluate systems responsibly. Draws on philosophy, social science, and domain expertise. May sit inside a technology company, a nonprofit, a university, or a government agency. AI Policy Analyst and Advisor: Works with governments, legislatures, regulatory agencies, and international bodies to develop AI governance frameworks. Requires understanding of both the technology and political and legal institutions. A field that barely existed a decade ago and is now urgently hiring. Domain Expert and AI Translator: A radiologist, lawyer, teacher, or financial analyst who deeply understands both their domain and AI capabilities, enabling them to specify what AI systems should do, evaluate whether they are doing it correctly, and communicate requirements across the technical-domain boundary. These roles are extraordinarily valuable and in short supply. UX Researcher and Designer for AI: Designs the human side of AI systems — how people interact with AI, how they understand what it can and cannot do, and how errors are communicated and corrected. Requires expertise in human-computer interaction, cognitive psychology, and design. AI Journalist and Science Communicator: Explains AI developments, capabilities, and implications to public audiences accurately and accessibly. Shapes public understanding, which in turn shapes political choices. Requires deep subject-matter knowledge and excellent communication skills. AI Lawyer and Compliance Specialist: Advises on liability, intellectual property, data protection, and regulatory compliance in AI deployment. As AI regulation matures, legal expertise specific to AI is increasingly valuable.

Choosing Your Path: What the Field Needs

One of the most important things to understand about the current state of AI careers is that the field is not uniformly over- or under-supplied. Certain skills are abundant relative to demand; others are severely scarce relative to need. There are many people who can train a neural network. There are very few people who can simultaneously understand a complex AI system technically, articulate its failure modes clearly to policymakers, and propose governance mechanisms that are both technically feasible and politically achievable. There are many ML engineers; there are very few AI policy specialists with genuine technical literacy. Similarly, domain expertise combined with AI literacy is extraordinarily rare and valuable. A cardiologist who understands the failure modes of AI-assisted diagnostic tools, and can evaluate whether a proposed deployment is clinically safe, is worth far more to a hospital system than a generic ML engineer. A teacher who understands both pedagogy and the limitations of AI tutoring tools can help schools deploy AI far more responsibly than technologists working alone. The field also urgently needs diversity of background, experience, and perspective. AI systems built by homogeneous teams tend to reflect the assumptions and blind spots of those teams. Including people who have experienced discrimination, who come from non-Western cultural contexts, or who have been on the receiving end of algorithmically-mediated decisions produces better systems — not for moral reasons alone, but for engineering reasons.