Technical Program Manager, Evals

About Anthropic

Anthropics 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.

About the Role

As a Technical Program Manager for model evaluations youll own end-to-end coordination of our evaluation ecosystem building a feedback loop from shaping eval strategy during early model development through launch execution. Youll be the critical bridge between Research Product Marketing and Engineering teams.

This role sits at the intersection of frontier AI research and product launches. Evals are an important part of how we measure whether our models meet the barfor capability safety and competitive positioning. Beyond launch coordination youll help scale our evals ecosystem: from early-stage model evals for RL environments to the systems and infrastructure on which evals run to tooling that enables the whole pipeline. A strong TPM in this space can immediately reduce chaos during launches while also driving systemic improvements that compound over time.

Key Responsibilities:

Launch Coordination

• Standardize how evaluation results are generated documented compared across model versions and communicated to stakeholders
• Own end-to-end eval readiness for model launchestracking which evals are ready which need scores on past models and which meet the bar for marketing materials
• Establish and enforce clear criteria for eval inclusion: scores on historical models state-of-the-art performance and competitor comparisons
• Coordinate between research teams marketing and product to consolidate eval status into a single source of truth
• Maintain a high bar: ensure reported statistics reflect model capabilities in an honest accurate and transparent way

Ecosystem Development

• Get involved early in model development cycles helping shape eval plans for RL environments
• Partner with research and infrastructure teams to improve underlying evals infrastructureeval-syncer reliability results storage and querying automation capabilities
• Drive prioritization of eval tooling enhancements based on researcher needs
• Identify patterns across launches and drive systemic fixes rather than point solutions
• Work with PMs and researchers to improve and implement high priority evals launches
• Maintain and prioritize the eval roadmapworking with cross-functional teams to identify which new evals are needed for upcoming launches and product requirements
• Implement an operating model that reflects an evals environment with increasing complexity

Process & Systems

• Build lightweight but rigorous tracking systemsmoving key information into structured formats that enable better decision-making
• Create eval dashboards that provide real-time visibility into training progress on hero evals enabling earlier intervention when scores look concerning
• Document eval processes requirements and lessons learned to build institutional knowledge
• Coordinate compute allocation for large-scale evals with infrastructure teams

You May Be a Good Fit If You:

• Have 5 years of technical program management experience with a track record of bringing order to chaotic high-stakes coordination problems
• Possess scientific depth and a very high quality bar for data
• Have experience with ML/AI evaluation methodologies benchmarking or research quality assurance
• Have a background in research operations scientific publishing or data quality management
• Have previous experience as data analyst data scientist or software engineer
• Can build trust with research teams by understanding their work deeply enough to add value beyond coordination
• Are skilled at cross-functional coordination involving research product marketing and engineeringnavigating competing priorities and driving alignment
• Have working familiarity with data analysis tools (SQL Python or similar) for querying eval results and building dashboards
• Have familiarity with LLM capabilities and limitations and experience working with AI research teams
• Excel at written and verbal communication translating technical nuance for marketing stakeholders while maintaining precision
• Thrive in unstructured environments with a bias toward action and a knack for creating clarity in ambiguous situations
• Have extremely high ownership and attention to detail

Deadline to apply:None applications will be received on a rolling basis.

Logistics

Education requirements: We require at least a Bachelors degree in a related field or equivalent experience.

Location-based hybrid policy: 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.

Visa sponsorship:We do sponsor visas! However we arent 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.

We encourage you to apply even if you do not believe you meet every single qualification. 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 youre interested in this work. We think AI systems like the ones were 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.

How were different

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. Were 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.

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.

Come work with us!

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. Guidance on Candidates AI Usage:Learn aboutour policyfor using AI in our application process