Note: We have a preference for candidates that are able to work out of NYC or SF for this role!

About Patronus AI

Patronus AI’s mission is to provide the security and risk management layer for AI. We are solving the problem of scalable oversight - how can humans continue to supervise AI systems when AI far outperforms them in many real world scenarios? Our vision is a world in which AI evaluates AI.Our founding team comes from top applied ML and research backgrounds, including Facebook AI Research (FAIR), Airbnb, Meta Reality Labs, and quant finance. As a team, we have published research papers at top ML conferences (NeurIPS, EMNLP, ACL), designed and launched Airbnb’s first conversational AI assistant, pioneered causal inference at Meta Reality Labs, exited a quant hedge fund backed by Mark Cuban, and scaled 0→1 products at high growth startups. We are backed by Lightspeed Venture Partners and high profile operators like Amjad Masad, Gokul Rajaram, and Fortune 500 executives and board members. We are advised by Douwe Kiela, Adjunct Professor at Stanford University and former Head of Research at Hugging Face.

Benefits

• Competitive salary and equity packages
• Health, dental, and vision insurance plans
• 401k plan
• Unlimited PTO
• Fun global offsites!

Responsibilities

As a Machine Learning Engineer at Patronus AI, you will solve the most important and challenging open research problems facing society’s adoption of AI today.

In this role, you will:

• Develop state-of-the-art systems for AI evaluation.
• Train language models for novel use cases, such as evaluating whether content is engaging, hallucinatory, age appropriate, or contains PII.
• Evaluate whether language models are aligned with human preferences.Collect high quality, novel datasets for classification and generative tasks, through synthetic data augmentation techniques and publicly available datasets.
• Conduct novel research on red teaming language models.
• Keep up to date on current literature and latest technologies, and synthesize findings in research updates.
• Experiment with latest technologies and proactively suggest experiments and improvements to systems.
• Maintain a research journal and well organized results logs.
• Benchmark performance of difference models by running robust experiments and ablations.
• Meet regularly with the CTO and research advisor to discuss progress updates and next steps.
• Develop efficient, scalable systems for hosting models for production use cases using AWS, Docker and Kubernetes.

Qualifications

“The number one qualification to succeed in this machine learning course is gumption.” - John Lafferty, CS Professor at Yale

Above all, we look for a proactive mindset, willingness to learn, relentless drive, and passion for research and engineering. You are a great fit if you have a background in the following:

• BS/MS in Computer Science, Mathematics, Statistics, or other quantitative field; PhD preferred
• Knowledge and understanding of state-of-the-art machine learning concepts, with a focus on NLP.
• Familiarity with transformer-based architectures, attention mechanisms, evaluation metrics and benchmarks
• Experience training language models in applied or research settings. Bonus if you have overcome challenges in training LLMs, scalable inference, or putting a LLM into production!
• Deep familiarity with pytorch and ML tooling such as job schedulers, the Hugging Face transformers library, and Weights & Biases.
• Experience conducting AI research in an academic or industry research lab. Bonus if you have published and reviewed papers in academic conferences (NeurIPS, ICML, ACL etc.).
• Ability to synthesize research papers and stay up-to-date on latest developments in research and open source.
• Have good character, integrity and respect for others!

Patronus AI is an equal opportunity employer. We celebrate diversity in our workplace, and all qualified applicants will receive consideration for employment without regard to age, ancestry, color, family or medical care leave, gender identity or expression, genetic information, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran status, race, religion, sex (including pregnancy), sexual orientation, or other legally protected characteristics.