The New AI-Era Job Titles, Explained: From Solutions Engineer to AI Strategist

Quick answer

The new AI-era job titles - AI solutions engineer, AI automation specialist, AI workflow designer, AI strategist, AI agent builder, AI implementation consultant, AI researcher, AI ethics specialist and technical founder - sound bewildering, but they cluster into a few simple families: people who build AI systems, people who decide where AI should be used, and people who keep it safe and honest. Almost all of them reward the same durable capabilities rather than a memorised tool: the ability to understand a problem, direct AI at it, evaluate whether the output can be trusted, and build something real. The titles are new and will keep changing; the World Economic Forum estimates 39% of workers' core skills will shift by 2030. The good news for a young person is that you do not chase the title - you build the judgement beneath it, and the titles become a menu rather than a mystery.

Why this matters now

These titles matter because they are where the labour market is actually moving, and because the noise around them is loud enough to mislead a careful family. A teenager who reads ten job ads for "AI strategist" or "AI agent builder" could be forgiven for thinking they are hopelessly behind, or that they need to master a dozen unrelated specialisms. Neither is true. The roles are real, but they are variations on a small number of underlying capabilities.

The macro picture is the reason the titles exist at all. The World Economic Forum's Future of Jobs Report 2025 projects that by 2030 some 170 million new roles will be created and 92 million displaced, a net gain of around 78 million. The fastest-growing roles in percentage terms are big-data specialists, fintech engineers, and AI and machine-learning specialists; the fastest-declining are clerical and secretarial - cashiers, administrative assistants, bank tellers and data-entry clerks. Fully 86% of employers told the WEF they expect AI to transform their business. The new titles are, in effect, names for the work on the growing side of that ledger.

The value, though, is not in adoption - it is in capability. McKinsey's State of AI 2025 found that 88% of organisations now use AI but only about 7% have fully scaled it, while PwC's 2025 Global AI Jobs Barometer found AI-skill roles carry a 56% wage premium. Together those numbers explain every title below: organisations have the technology and cannot extract value from it, so they pay a premium for people who can - by building, deciding or governing.

Australia sits inside this shift, not outside it. The Tech Council of Australia, supporting the federal target of 1.2 million tech workers by 2030, reports the workforce stood at around 950,000 in mid-2025 - about 650,000 short - with tech vacancy rates roughly 60% above the national average, and warns, drawing on Australian Computer Society analysis, that the trajectory is "not on track". Most of the growth is "indirect tech": these roles increasingly sit inside banks, retailers, miners and government, not just software companies. Jobs and Skills Australia's 2025 Our Gen AI Transition report adds the reassuring finding that gen AI augments more work than it replaces and lifts demand for human skills. The new titles are how that demand is advertised.

What the new AI-era job titles really mean

Here is the plain-English version of each, with its family. Two - the forward deployed engineer and the AI product manager - have their own deep guides, linked below.

• AI solutions engineer (builder). Builds working AI solutions for customers, bridging a general platform and a client's specific problem - writing code and integrating AI into their systems. It overlaps heavily with the forward deployed engineer, the role The Pragmatic Engineer documents as originating at Palantir and now hired by OpenAI, Anthropic and Google. Core skills: engineering plus customer communication.
• AI automation specialist (builder). Designs systems that use AI to do repetitive work automatically - sorting, drafting, moving data between tools - with human oversight. As the WEF flags clerical roles among the fastest-declining, the people who build the automations are in rising demand. Core skills: process thinking plus practical tool fluency.
• AI workflow designer (builder/decider). Maps how work flows through a team and redesigns it so AI handles the right steps and humans keep the rest. A close cousin to the automation specialist, but focused on the whole process rather than a single task. Core skills: systems thinking plus judgement about what to automate.
• AI agent builder (builder). Creates AI systems that take multi-step actions on their own - researching, drafting and filing a task rather than just answering. Designs how the agent reasons, what tools it can use, and the guardrails that keep it safe. Core skills: engineering plus rigorous evaluation, because an agent that acts can fail in more ways than one that only talks.
• AI strategist (decider). Helps an organisation decide where AI is worth using and where it is not, weighing cost, risk and ethics. This is the role most directly aimed at McKinsey's 7%-scaled problem. Core skills: analytical thinking plus commercial and ethical judgement.
• AI implementation consultant (decider/builder). Guides an organisation through adopting AI - choosing tools, redesigning processes, training people, managing the change. Part strategist, part hands-on. Core skills: communication, project judgement and enough technical fluency to be credible.
• AI researcher (builder, specialist). Works on improving the underlying AI itself - new methods, models and techniques. The most technical and credential-heavy role, usually requiring advanced study. It is not the same as the applied roles above. Core skills: deep mathematics, computer science and experimentation.
• AI ethics specialist (governor). Makes sure AI systems are fair, safe, transparent and honest - assessing bias, privacy and harm, guided by frameworks like the OECD AI Principles (2019, updated 2024). A growing role as regulation tightens. Core skills: ethical reasoning, social awareness and enough technical understanding to assess real systems.
• Technical founder (builder + decider). Starts a company and can build the product, at least initially. AI tools now let a small team build real software fast, putting this within closer reach. Core skills: engineering, product judgement and the resolve to solve a real problem.

What these share is more revealing than what separates them. None is won by memorising prompts; each rewards some mix of building, deciding and governing, and underneath all three sits judgement. That is why UNESCO's AI Competency Framework for Students (2024) sequences competencies across a human-centred mindset, ethics, AI techniques and AI system design rather than around any tool: the framework is describing the capabilities these jobs actually pay for.

A simple comparison

The titles are easier to hold as three families. This table is a map, not a ranking - and the same person often moves between columns over a career.

The pattern is hard to miss. Almost every column is marked "high" on judgement, while the coding requirement varies enormously. A young person who concludes they must learn one specific tool to enter this world has misread the table. The asset that opens the most columns is judgement - the capacity to understand a problem, direct AI at it, and evaluate whether the result can be trusted.

How a young person gets toward any of these

Because the roles share a foundation, a teenager does not have to pick one at fifteen. They build the common base, and the doors open later - the same base we argue every student should carry in the AI skills students need before they leave school.

1. Build real things. Every builder role, and most decider roles, are won by people with evidence they can make something work. PBLWorks' research on Gold Standard project-based learning is clear that durable capability comes from authentic artefacts, not theory.
2. Practise deciding and evaluating. The decider and governor roles run on judgement - what to build, whether it is good, whether it is fair. A student who habitually asks "should this exist, and how would I know if it worked?" is rehearsing half the table.
3. Learn to communicate technical work plainly. Solutions engineers, consultants and founders all live or die on this. Harvard Project Zero's framing of understanding as a flexible performance - using knowledge in new situations, including explaining it - captures the skill exactly.
4. Keep ethics in view. The AI ethics specialist is a role, but ethical judgement is part of every role on the list. The OECD AI Principles are a sound place for a curious teenager to start.
5. Stay in command of the tools. The single thread through all of it: direct AI deliberately and verify what it returns, rather than deferring to it.

This is the Edison Method in one sentence - Understand, Use, Evaluate, Build, Lead - and it is deliberately title-agnostic. Build those five capabilities and a student is positioned for the builder, decider and governor roles alike.

Three starter projects that build toward several of these at once

Each project deliberately exercises more than one family, so a teenager keeps their options open.

• The automation that saves someone an hour. What the student does today: a teenager finds one repetitive task a family member or club does weekly and builds a small AI-assisted process to ease it. How AI assists: it handles the repetitive drafting or sorting. What the human must verify: that the person can trust, understand and override it. The learning outcome: the automation-specialist and workflow-designer mindset - seeing work as a process to improve. The control: the human stays in charge of every consequential step.
• The "should we build this?" memo. What the student does today: they pick a problem at their school and write a one-page memo arguing whether AI is the right tool for it, what it would cost in effort, and what could go wrong. How AI assists: it can surface counter-arguments. What the human must verify: that the reasoning and the recommendation are genuinely theirs. The learning outcome: the strategist and ethics-specialist habit of deciding whether, not just how. The control: the verdict is the student's.
• The mini-agent with guardrails. What the student does today: using a no-code or low-code tool, they build a simple multi-step helper - say, one that drafts a reply and flags anything it is unsure about - and then deliberately try to break it. How AI assists: it performs the steps. What the human must verify: where it fails, and what guardrail would prevent that. The learning outcome: the agent-builder's most important instinct, that an acting system needs rigorous evaluation. The control: the student decides what the agent is not allowed to do.

Common mistakes and misconceptions

These traps are common, and each one points a young person the wrong way.

• Chasing the title instead of the capability. Titles are unstable - the WEF's estimate that 39% of core skills will shift by 2030 applies to job names too. Build the judgement; let the title follow.
• Assuming every role needs deep coding. The table shows it does not. AI strategist, AI product manager and AI ethics specialist run primarily on judgement.
• Confusing the applied roles with AI research. AI researcher is a distinct, credential-heavy, mathematics-deep career. Most of the others are about using AI well, not inventing it.
• Believing the roles are only at tech companies. The Tech Council's data shows most growth is "indirect tech" - inside banks, retailers, government and mining. These jobs are spreading across the whole economy.
• Thinking it is all prompting. Prompting is a thin slice. The substance - across every family - is building, deciding and governing, none of which a memorised phrase can do. This is exactly why the question of whether prompt engineering is still a job matters: the skill is real, but it is becoming part of every role rather than a job of its own.

Where to go deeper

Two of these roles reward a closer look, because they are among the most significant and the most accessible to grow into. The forward deployed engineer guide unpacks the builder archetype - the engineer embedded in a customer's team, turning a platform into a solution. The AI product manager guide unpacks the decider archetype - the person who chooses what to build and whether it is good enough to ship, often with no coding at all. Together they show the two halves of how AI actually creates value: someone builds it, and someone decides it is worth building.

The recommendation is simple and a little contrarian. Do not start by picking a title from this list - most of them did not exist five years ago, and the list will look different in five more. Start by building the capability that sits under all of them: understand, direct, evaluate, build. Make real things, decide whether they should exist, and learn to explain them to a human. Do that through the teenage years and these titles stop being a source of anxiety and become what they should be - a menu. The names keep changing; the judgement is the thing that lasts.