Principal Data Scientist

About the Role

We're building the systems that turn one of the world's largest scientific corporations into research intelligence. That means production NLP pipelines running over millions of journal articles, extracting entities, classifications, claim tuples, and summaries optimized for use by downstream agentic applications. We're looking for a principal data scientist to own domain-specific content modeling work end to end, from the eval set through the pipeline stage that ships it. You'll join a small, senior team where data scientists own their models in production. You'll write the code, own the evaluations, ship the changes, and stay accountable for the outcomes. This is a hands-on role for someone who wants to see their models through to real users in a rapidly evolving market.

Responsibilities

• Design and build NLP enrichment pipelines that extract entities, classifications, claims, and summaries from scientific full-text at scale.
• Compare NLP approaches to extraction and enrichment against LLM-based approaches, and pick the right tool for each task. This includes traditional NLP (NER, sequence labeling, classification), embedding-based retrieval, LLM prompting, and fine-tuned smaller models.
• Own evaluation: build golden sets in consultation with SMEs and vendors, choose metrics, and balance speed, quality, and cost.
• Write production-quality Python, manage concurrency and cost for high-volume LLM workloads, and structure code for engineers and other data scientists.
• Collaborate with data engineers on pipeline orchestration using tools like Airflow and Dagster, ensuring idempotent, retryable, and evaluable stages.
• Contribute to agentic AI applications, shaping how agents reason over the enriched corpus and ground their answers.
• Work directly with editors, product managers, and engineers to bring the modeling perspective into product decisions and translate feedback into modeling work.

Requirements

• Deep Python expertise with production-scale experience, including concurrency management (asyncio, threads, queues).
• Strong NLP background in both modern (LLMs, transformers, embeddings) and classical (NER, classification, sequence labeling) approaches.
• Proven ability to compare approaches, defend choices with evaluations and cost estimates, and handle performance drift.
• Track record of shipping systems that deliver real-user value, not just prototypes or papers.

Qualifications

• Experience working with scientific or scholarly text (preferred).
• Familiarity with AWS (S3, Batch, Lambda, SageMaker) and data lake patterns (Parquet, Iceberg) (preferred).
• Experience running LLMs in production under cost and latency constraints (preferred).
• Exposure to agentic AI applications (tool use, multi-step reasoning, guardrails, trajectory evaluation) (preferred).

Pay

Salary Range: 140,000.00 USD to 200,733.33 USD. The ultimate salary will be set based on factors including geographic location, skills, and competencies.