AI Researcher · Designs and tests the algorithms behind machine intelligence, in a field now racing to automate a growing share of its own research process.
Darker cells mean a higher score for this topic on that metric.
LessMore
Last reviewed Sources & creditsMedia creditsMethodology
Most days split between reading recent papers, writing and debugging training code, waiting on and analyzing results from long-running experiments, and discussing findings with collaborators. Contrary to the popular image, a large share of the job is diagnosing why a model is underperforming or a result won't replicate, not brainstorming new architectures from scratch.
For an independent research role at a university or a top industry lab, effectively yes — a PhD remains the field's real credential, since there is no licensing exam. Strong engineers without one increasingly enter through industry residency programs or a public record of open-source work and cited preprints, but titles like 'research scientist' still skew heavily toward doctorate holders.
Parts of it clearly are: literature summarizing, boilerplate experiment code and first-draft related-work sections are already routinely AI-assisted, and labs are openly experimenting with systems that propose and run their own experiments. Framing what a result actually means, and taking responsibility for a published claim, have proven much harder to hand over.
It varies enormously by employer and seniority. The US median for the broader 'computer and information research scientist' category was $140,910 in 2024, but new PhDs joining a large technology company's research lab often start well above that, and a small number of senior researchers at frontier AI labs have reportedly been offered total packages worth seven or eight figures during the 2024–2025 hiring competition.
The lines blur constantly, but a rough distinction holds: a researcher's output is usually a paper, a new method or an insight about why something works, judged by peer review; an engineer's output is usually a working system in production, judged by whether it ships and holds up under real use. Many people do both across a career.
Python dominates, paired with a deep-learning framework — PyTorch is now the most common in research, with JAX popular for large-scale and Google-affiliated work. Beyond code, researchers depend on shared GPU or TPU compute clusters, experiment-tracking tools like Weights & Biases, and arXiv, where most new results circulate as preprints months before formal peer review.
Popular culture pictured artificial intelligence long before the field itself made much progress, and the gap between the two has shaped how the public reacts to every real advance since. A menacing or godlike computer was a stock character in film and fiction well before researchers had built anything more capable than a program that could play tic-tac-toe.
Inside the field, the culture runs differently: sardonic in-jokes about hype cycles, deadline-driven conference rituals, and an unusually blunt tradition — inherited partly from academia, partly from open-source software — of publishing failures and negative results alongside successes, at least in principle.
How much status the profession carried in each era, on a 0–100 scale.
Early optimism made AI researchers minor scientific celebrities within computing, buoyed by confident public predictions that human-level machine intelligence was decades, not centuries, away.
The first AI winter and its funding cuts pushed 'artificial intelligence' out of favor as a term; many researchers rebranded their work as 'expert systems,' 'pattern recognition' or plain 'computer science' to keep grants flowing.
A brief commercial revival — corporations paying real money for rule-based expert systems like DEC's XCON — restored some prestige and funding before the approach's limits became clear again by the decade's end.
Through the second AI winter and its long aftermath, most of the field's steady progress happened under other names — machine learning, statistics, data mining — with little public attention paid to it at all.
Deep learning's public breakthroughs, billion-dollar funding rounds, a Nobel Prize and a Turing Award have made leading AI researchers some of the most publicly recognized scientists alive, alongside real public anxiety about what they are building.
HAL 9000, a spacecraft's calm, murderous computer, became the default cultural shorthand for an AI that outgrows its instructions — a reference researchers still hear more than any other, half a century on.
A military supercomputer nearly triggers nuclear war after being asked to 'play' a game with real missiles, dramatizing Cold War anxieties about autonomous systems that would resurface decades later around AI safety.
Frames a private test of machine consciousness as a variation on Turing's imitation game, and became a reference point in real debates over whether large language models can be meaningfully said to 'understand' anything.
Follows DeepMind's program through its 2016 match against Go champion Lee Sedol, capturing the research team's own surprise at 'move 37' — one of the few widely seen films showing what an actual research lab's work looks like.
A Japanese cyberpunk landmark exploring machine consciousness and identity years before Western films caught up to the same questions, and a major influence on later American science fiction about AI.
A US crime drama built around a mass-surveillance AI predicting violent crimes before they happen, popularizing questions about algorithmic prediction and bias years before they became mainstream policy debates.
NeurIPS, ICML and ICLR submission deadlines fall at a fixed hour — traditionally 11:59 p.m. 'Anywhere on Earth' — and trigger a predictable ritual of last-minute experiments, all-night writing and a rush of new preprints appearing on arXiv in the final hours.
Accepted authors stand for hours beside a printed poster in a crowded hall, repeating their pitch to a rotating stream of strangers — a rite of passage every researcher remembers, for better or worse, from their first accepted paper.
Publishing a paper alongside code, trained weights and a public benchmark result has become an expected package rather than an optional extra, turning a lab's internal milestone into a reproducibility test the whole field can immediately run.
None of this settles the field's oldest argument: whether what these systems do is a real step toward intelligence or an increasingly convincing form of pattern-matching. Researchers themselves are split, often within the same lab.
What has stayed constant is the mismatch between the image and the work: neither HAL 9000 nor a chrome-plated robot resembles a researcher squinting at a loss curve that refuses to go down at 1 a.m. before a deadline.
Closest neighbours on the six-score profile — not the same field only.
Designs and fabricates the transistors inside every computer, phone and weapon, using machines precise enough that only a few factories on Earth can run them.
AI-resistant 60 📦Builds the systems that train, deploy, monitor and govern machine-learning models in production.
AI-resistant 54 🩺Diagnoses illness and manages health for years afterward through examination and evidence, not a single operation — medicine's generalist and long-term guide.
AI-resistant 67 🧠Diagnoses and treats mental illness through medication and therapy, one of medicine's only specialties with legal authority to detain a patient in crisis.
AI-resistant 73 👔Advises clients, drafts the documents that bind them, and argues their case when it reaches court — carrying personal legal liability if the advice is wrong.
AI-resistant 58 🔐Protects systems, data and people by finding, preventing and responding to digital attacks.
AI-resistant 63Writes, tests and maintains the code that runs modern life — and is one of the first professions watching AI automate its own daily work.
AI-resistant 35 🛰️Designs, analyzes and certifies the aircraft, rockets and spacecraft that leave the ground, working to safety margins that leave no room for guessing.
AI-resistant 74 🌉The profession that turns rivers, rock and gravity into bridges, roads and clean water — civilization's quiet load-bearing trade since Imhotep.
AI-resistant 72 🔌Designs and fabricates the transistors inside every computer, phone and weapon, using machines precise enough that only a few factories on Earth can run them.
AI-resistant 60 🦾Designs the machines that sense, decide and act in the physical world, where the hard problem was never intelligence but the world itself.
AI-resistant 65 🔐Protects systems, data and people by finding, preventing and responding to digital attacks.
AI-resistant 63 📦Builds the systems that train, deploy, monitor and govern machine-learning models in production.
AI-resistant 54 🌬️Designs, builds and improves wind, solar, storage and grid systems that turn renewable resources into dependable electricity.
AI-resistant 72