Member of Technical Staff

Societies; City Of London; On-site; GBP 90,000 - 150,000; Full time

Job summary

Societies in London (Holborn) seeks a proactive Member of Technical Staff to advance frontier societal foundation models across training, evaluation, and deployment. You will design methods to train models from real-world observations and reimagine concepts like auto-regression and diffusion to capture human decision-making.

You will also create robust evaluation that tells us how well simulations match reality and work directly with our CEO and growing research team, alongside product and

Qualifications

• 5+ years of experience in an applied research or research engineering role.
• Experience in post-training and evaluating LLMs: RL, dataset and reward design, synthetic data.
• Proven track record deploying trained models commercially where they generated revenue.
• Research literacy to navigate and push forward the frontier of ML research and its applications in behavioural science.
• Python proficiency; our simulation engine is Python-based.

Responsibilities

• Train large-scale models using frontier ML methods to simulate individuals, subgroups, and populations.
• Ground models in data by gathering and learning from large-scale behavioural datasets.
• Own evaluation and calibration methodology; design evals, benchmarks, and validation studies.
• Run experiments end-to-end: hypothesis, infrastructure, training, analysis, decision.
• Push the frontier by writing and publishing academic content on synthetic research.

Skills

Applied research

Research engineering

Python

LLMs evaluation

RL

Job description

The problem

We simulate societies: how thousands of specific individuals influence each other and form opinions across a multiverse of scenarios.

The combinatorics are as hard as protein folding. Give 1,000 people a for-or-against position on 10 issues, and there are 10 3,000 possible group opinion states. DeepMind only has to model 10 300 ways a protein can fold. Searching that space fast enough to provide answers in hours, not months, is a frontier machine learning problem.

To solve this, we are building a Societal World Model: an AlphaFold for societies. We are already informing strategies for clients with $3T+ in combined market cap. For a few seconds during one client simulation, we were 2% of Google’s global AI throughput.

About the role

We’re looking for a Member of Technical Staff to advance the frontier of our societal foundation models, working across the stack to improve their training, evaluation, and deployment.

You’ll design innovative methods to efficiently train models from real-world observations, reimagine how concepts like auto-regression and diffusion can capture human decision-making, and create the evaluation methodology that tells us how well our simulations match reality. You’ll work directly with our CEO and our growing research team, while partnering closely with product and forward-deployed engineers to move ideas from a research notebook into production.

This role is for someone who wants to do frontier research at the intersection of machine learning and social science and is motivated by applying their work to the world’s most consequential decisions.

What you’ll do

• Train large-scale models. Apply frontier ML methods in supervised learning and natural language processing to develop state-of-the-art simulations of individuals, subgroups, and populations.
• Ground models in data. Develop innovative methods to gather and learn from large-scale behavioural datasets.
• Own evaluation and calibration methodology. Design the evals, benchmarks, and validation studies that measure simulation fidelity against real-world decisions, and make the results interpretable to the team.
• Run experiments end-to-end. Hypothesis, infrastructure, training run, analysis, decision. Build the experimentation infrastructure that lets us iterate fast and trust the numbers.
• Push the frontier. Write, critique, and publish academic content to frame how the world understands and evaluates synthetic research.

What we’re looking for

• 5+ years of experience in an applied research or research engineering role.
• Experience in post-training and evaluating LLMs: RL, dataset and reward design, synthetic data, or related training methods.
• Demonstrated track record of success in deploying trained models in a commercial setting where they generated revenue.
• Research literacy to navigate and push forward the frontier of ML research and its applications in behavioural science.
• Python-strong: our simulation engine is Python-based, and this role lives in Python day-to-day.

Nice to have

• Interdisciplinary experience in behavioural science, computational social science, psychometrics, statistics, complexity science, or a related empirical field.
• AI-pilled: motivated to foster an environment where AI agents make meaningful contributions across the research, product-engineering, and commercial deployment stack.
• Experience scaling an ML research team in an industry setting.

Logistics

• Location: London (Holborn), in-office. We build better together, and this role is no exception.
• Salary: Competitive with lab market rates.

Why join now

If you want to work on something truly frontier, this is it. You’ll be working directly with some of the smartest people on the planet, on one of the hardest problems humanity has yet to solve. In a world on the brink of exponential, recursive change, there has never been a more important time to help society understand itself.