Research Engineer / Scientist, Alignment Science

Other Jobs To Apply

<strong><strong>About Anthropic</strong></strong><br><br>Anthropic's mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.<br><br><strong>About the role:</strong><br><br>You want to build and run elegant and thorough machine learning experiments to help us understand and steer the behavior of powerful AI systems. You care about making AI helpful, honest, and harmless, and are interested in the ways that this could be challenging in the context of human-level capabilities. You could describe yourself as both a scientist and an engineer. As a Research Engineer on Alignment Science, you'll contribute to exploratory experimental research on AI safety, with a focus on risks from powerful future systems (like those we would designate as ASL-3 or ASL-4 under our Responsible Scaling Policy), often in collaboration with other teams including Interpretability, Fine-Tuning, and the Frontier Red Team.<br><br>Our blog provides an overview of topics that the Alignment Science team is either currently exploring or has previously explored. Our current topics of focus include...<ul><li><strong>Scalable Oversight: </strong>Developing techniques to keep highly capable models helpful and honest, even as they surpass human-level intelligence in various domains.</li><li><strong>AI Control: </strong>Creating methods to ensure advanced AI systems remain safe and harmless in unfamiliar or adversarial scenarios.</li><li><strong>Alignment Stress-testing</strong><strong>:</strong> Creating model organisms of misalignment to improve our empirical understanding of how alignment failures might arise.</li><li><strong>Automated Alignment Research: </strong>Building and aligning a system that can speed up & improve alignment research.</li></ul><br><br>Note: Currently, the team has a preference for candidates who are able to be based in the Bay Area. However, we remain open to any candidate who can travel 25% to the Bay Area.<br><br><strong>Representative projects:</strong><br><ul><li>Testing the robustness of our safety techniques by training language models to subvert our safety techniques, and seeing how effective they are at subverting our interventions.</li><li>Run multi-agent reinforcement learning experiments to test out techniques like AI Debate.</li><li>Build tooling to efficiently evaluate the effectiveness of novel LLM-generated jailbreaks.</li><li>Write scripts and prompts to efficiently produce evaluation questions to test models' reasoning abilities in safety-relevant contexts.</li><li>Contribute ideas, figures, and writing to research papers, blog posts, and talks.</li><li>Run experiments that feed into key AI safety efforts at Anthropic, like the design and implementation of our Responsible Scaling Policy.</li></ul><br><br><strong>You may be a good fit if you:</strong><br><ul><li>Have significant software, ML, or research engineering experience</li><li>Have some experience contributing to empirical AI research projects</li><li>Have some familiarity with technical AI safety research</li><li>Prefer fast-moving collaborative projects to extensive solo efforts</li><li>Pick up slack, even if it goes outside your job description</li><li>Care about the impacts of AI</li></ul><br><br><strong>Strong candidates may also:</strong><br><ul><li>Have experience authoring research papers in machine learning, NLP, or AI safety</li><li>Have experience with LLMs</li><li>Have experience with reinforcement learning</li><li>Have experience with Kubernetes clusters and complex shared codebases</li></ul><br><br><strong>Candidates need not have:</strong><br><ul><li>100% of the skills needed to perform the job</li><li>Formal certifications or education credentials</li></ul><br><br>The expected salary range for this position is:<br><br>Annual Salary:<br><br>$280,000-$690,000 USD<br><br><strong><strong>Logistics</strong></strong><br><br><strong>Education requirements: </strong>We require at least a Bachelor's degree in a related field or equivalent experience.<strong>Location-based hybrid policy:</strong> Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.<br><br><strong>Visa sponsorship:</strong> We do sponsor visas! However, we aren't able to successfully sponsor visas for every role and every candidate. But if we make you an offer, we will make every reasonable effort to get you a visa, and we retain an immigration lawyer to help with this.<br><br><strong>We encourage you to apply even if you do not believe you meet every single qualification.</strong> Not all strong candidates will meet every single qualification as listed. Research shows that people who identify as being from underrepresented groups are more prone to experiencing imposter syndrome and doubting the strength of their candidacy, so we urge you not to exclude yourself prematurely and to submit an application if you're interested in this work. We think AI systems like the ones we're building have enormous social and ethical implications. We think this makes representation even more important, and we strive to include a range of diverse perspectives on our team.<br><strong><strong>How we're different</strong></strong><br><br>We believe that the highest-impact AI research will be big science. At Anthropic we work as a single cohesive team on just a few large-scale research efforts. And we value impact - advancing our long-term goals of steerable, trustworthy AI - rather than work on smaller and more specific puzzles. We view AI research as an empirical science, which has as much in common with physics and biology as with traditional efforts in computer science. We're an extremely collaborative group, and we host frequent research discussions to ensure that we are pursuing the highest-impact work at any given time. As such, we greatly value communication skills.<br><br>The easiest way to understand our research directions is to read our recent research. This research continues many of the directions our team worked on prior to Anthropic, including: GPT-3, Circuit-Based Interpretability, Multimodal Neurons, Scaling Laws, AI & Compute, Concrete Problems in AI Safety, and Learning from Human Preferences.<br><strong><strong>Come work with us!</strong></strong><br><br>Anthropic is a public benefit corporation headquartered in San Francisco. We offer competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and a lovely office space in which to collaborate with colleagues.

Back to blog

Common Interview Questions And Answers

1. HOW DO YOU PLAN YOUR DAY?

This is what this question poses: When do you focus and start working seriously? What are the hours you work optimally? Are you a night owl? A morning bird? Remote teams can be made up of people working on different shifts and around the world, so you won't necessarily be stuck in the 9-5 schedule if it's not for you...

2. HOW DO YOU USE THE DIFFERENT COMMUNICATION TOOLS IN DIFFERENT SITUATIONS?

When you're working on a remote team, there's no way to chat in the hallway between meetings or catch up on the latest project during an office carpool. Therefore, virtual communication will be absolutely essential to get your work done...

3. WHAT IS "WORKING REMOTE" REALLY FOR YOU?

Many people want to work remotely because of the flexibility it allows. You can work anywhere and at any time of the day...

4. WHAT DO YOU NEED IN YOUR PHYSICAL WORKSPACE TO SUCCEED IN YOUR WORK?

With this question, companies are looking to see what equipment they may need to provide you with and to verify how aware you are of what remote working could mean for you physically and logistically...

5. HOW DO YOU PROCESS INFORMATION?

Several years ago, I was working in a team to plan a big event. My supervisor made us all work as a team before the big day. One of our activities has been to find out how each of us processes information...

6. HOW DO YOU MANAGE THE CALENDAR AND THE PROGRAM? WHICH APPLICATIONS / SYSTEM DO YOU USE?

Or you may receive even more specific questions, such as: What's on your calendar? Do you plan blocks of time to do certain types of work? Do you have an open calendar that everyone can see?...

7. HOW DO YOU ORGANIZE FILES, LINKS, AND TABS ON YOUR COMPUTER?

Just like your schedule, how you track files and other information is very important. After all, everything is digital!...

8. HOW TO PRIORITIZE WORK?

The day I watched Marie Forleo's film separating the important from the urgent, my life changed. Not all remote jobs start fast, but most of them are...

9. HOW DO YOU PREPARE FOR A MEETING AND PREPARE A MEETING? WHAT DO YOU SEE HAPPENING DURING THE MEETING?

Just as communication is essential when working remotely, so is organization. Because you won't have those opportunities in the elevator or a casual conversation in the lunchroom, you should take advantage of the little time you have in a video or phone conference...

10. HOW DO YOU USE TECHNOLOGY ON A DAILY BASIS, IN YOUR WORK AND FOR YOUR PLEASURE?

This is a great question because it shows your comfort level with technology, which is very important for a remote worker because you will be working with technology over time...