Research Engineer

As Atla’s Research Engineer, you'll develop and fine-tune language models as evaluators as well as use your insights to construct safety guardrails for client LLMs, pushing the state of the art further. As part of your role, you will:

• Steer LLMs to become strong evaluators aligned with human preferences using advanced post-training techniques.
• Curate and synthetically generate large-scale, high-quality datasets to refine our evaluator for assessing frontier and next-generation models.
• Develop comprehensive evaluation frameworks, including tooling, datasets and metrics for rigorous assessment of alignment and safety risks.
• Shape Atla’s evolving AI safety research and contribute significant findings to top-tier conferences in the field.
• Build and grow our applied research organisation, setting a high bar of excellence that propels Atla forward.

‍

Please note that this role is in-person (we can sponsor visas and offer international relocation support as a UK AI Futures partner)

Evidence of exceptional research engineering ability:

• Proficiency in training and evaluating language models across GPUs, preferably in PyTorch.
• Proven track record in empirical research, including designing and executing experiments, and effectively writing up and communicating findings.
• Strong software, ML, or research engineering experience with a focus on building robust, scalable system.
• Aptitude for distilling and applying ideas from complex research papers in our products.
• Experience at elite AI research labs (OpenAI, DeepMind, Meta, Anthropic, etc.).

• Publications in top AI conferences.
• Experience as an early engineer at a fast growing startup.
• Interested in and thoughtful about the impacts of AI technology.

You'll work by and thrive through our core principles:

Own the Outcome

• Create real value: Every action should deliver tangible, meaningful value for the people who use what we build.
• Drive to completion: Do the second 90%.
• Do fewer things, better: Prioritize focus over breadth.

Back the Team

• Collaborate for excellence: The whole is greater than the sum of its parts.
• Seek truth: Let the best ideas win, no matter where they come from, and let go of ego.
• Argue passionately, then commit fully: Debate fiercely, but once a decision is made, own it like it’s yours.

Drive the Mission

• Advance AI safety: Every action should contribute towards the safe development of AI.
• Go big or go home: “The people who are crazy enough to think they can change the world are the ones who do.”

Compensation

• £100K - £300K GBP
• Significant stake in equity as one of the early joiners
• Pension plan
• Medical, dental, and vision benefits

‍

“This role is supported by AI Futures Grants, a UK Government scheme designed to help the next generation of AI leaders meet the costs of relocating to the UK. AI Futures Grants provide financial support to reimburse relocation costs such as visa fees, the immigration health surcharge and travel/subsistence expenses. Successful candidates for this role may be able to get up to £10,000 to meet relocations costs, subject to terms and conditions.”