Engineering Manager, Research Data Platform

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. About the role Anthropic's researchers generate and depend on enormous amounts of data — training runs, evaluations, RL transcripts, annotations etc... The Research Data Platform team builds the systems that make that data easy to produce, find, query, and trust. We work in two modes: we build platform components that other systems plug into (for example, a metrics library that training frameworks integrate to record and retrieve run data), and we own core datasets end to end (for example, the data pipeline behind RL transcripts). As the team's tech lead, your job starts with our users. You'll work directly with researchers — and with the engineers who support them — to understand how they actually work, where managing data slows them down, and where a well-built platform component or a well-curated dataset would change what's possible. You'll turn what you learn into technical direction for the team, in partnership with the team's manager, who owns priorities and people. A central ambition you'll drive: a small set of canonical, well-documented datasets — starting with the core data model for RL — that researchers trust and standardize on, rather than every team managing its own copies. You'll spend your first few months close to the code and close to users: shipping improvements in our core systems, embedding with research teams, and building your own map of their workflows. As the team grows, this role has a natural path into formal people leadership for someone who wants it. Responsibilities · Work directly with researchers and the engineers supporting them to understand their workflows, identify the highest-leverage opportunities, and shape what the team builds next · Set the technical direction for the team across our platform and our datasets · Design and build platform components that other teams plug into — libraries, services, and interfaces such as the metrics library used by training frameworks · Own core datasets end to end: the pipelines that produce them, the schemas that define them, and the documentation and guarantees that make researchers trust them · Drive convergence toward canonical datasets — including the core data model for RL transcripts — that research teams standardize on · Lead complex, multi-quarter projects that span several systems and teams, staying hands-on in the code · Raise the team's technical bar through design reviews, mentorship, and the quality of your own work You may be a good fit if you: · Have built and operated data-intensive systems at scale — pipelines, storage layers, query systems — with strong instincts for data modeling and schema design that hold up as usage grows · Have set technical direction for a team, or owned the architecture of a data platform that other teams build on · Treat internal users as customers: you do the discovery work, iterate with users, and measure success by adoption rather than by shipping · Understand that researchers aren’t typical internal customers — the work is exploratory by nature, workflows differ from team to team, and requirements are discovered through experiments rather than specified up front · Can build for that motion — keeping interfaces stable and data trustworthy while use cases change underneath you, and judging when a quick, disposable solution serves research better than a durable one · Lead through influence — aligning engineers and stakeholders without relying on formal authority · Are results-oriented and pragmatic, willing to do unglamorous work when it's the highest-leverage thing · Are excited about learning the fundamentals of machine learning research (deep ML expertise is not required) · Care about the societal impacts of your work Strong candidates may also have · Experience with large-scale ETL and columnar or analytical storage (e.g., Spark, BigQuery, ClickHouse, DuckDB, Parquet) · Experience with metrics or experiment-tracking systems, or high-volume time-series data · Experience with dataset management, cataloging, or lineage tooling · Built developer tooling or internal data platforms for demanding technical users — including in domains like quantitative trading, where fast-moving, exploratory data work looks a lot like research · A working knowledge of machine learning · Worked in, or closely with, an ML research lab · Interest in — or experience with — people management and growing engineers For sales roles, the range provided is the role’s On Target Earnings ("OTE") range, meaning that the range includes both the sales commissions/sales bonuses target and annual base salary for the role. Annual Salary: $405,000—$850,000 USD Logistics Minimum education: Bachelor’s degree or an equivalent combination of education, training, and/or experience Required field of study: A field relevant to the role as demonstrated through coursework, training, or professional experience Minimum years of experience: Years of experience required will correlate with the internal job level requirements for the position 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. 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.