Note: The job is a remote job and is open to candidates in USA. Wiley is a scientific, education, and publishing company focused on transforming knowledge into impact. The Principal Data Scientist will build and own production NLP enrichment pipelines for scientific content, including model evaluation, implementation, and deployment. The role also contributes to agentic AI applications and collaborates with data engineers, editors, product managers, and engineers.
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. That means putting traditional NLP (NER, sequence labeling, classification), embedding-based retrieval, LLM prompting, and fine-tuned smaller models on the same table, and defending each choice with evaluation, cost, and operational tradeoffs. This is a core part of the job, not an occasional exercise
- Own evaluation. Build the golden sets in consultation with SMEs and vendors, choose the metrics, and make productive tradeoffs between speed, quality, and cost
- Write production-quality Python. Manage concurrency and cost for high-volume LLM workloads. Structure code that engineers can ship and other data scientists can extend
- Collaborate with a team of data engineers to orchestrate work in data pipeline and data build tools like Airflow and Dagster. Design idempotent, retryable, evaluable pipeline stages that stay reliable when a run fails at scale
- Contribute to agentic AI application work: tool-using systems that reason over the enriched corpus, where your NLP and evaluation background will shape how the agent grounds and defends its answers
- Work directly with editors, product managers, and engineers. Bring the modeling perspective into product decisions, and translate stakeholder pushback into concrete modeling work
Skills
- * Deep Python. You've written it in production, at scale, for years. You know when to reach for asyncio versus threads versus a queue, and you can explain the tradeoff clearly
- * Strong NLP background across modern (LLMs, transformers, embeddings, retrieval) and classical (NER, classification, sequence labeling) approaches. You've built evaluations and learned from the results
- * A habit of comparing approaches and choosing the right one for the task. You can defend "prompt a large LLM" and "train a small classifier on 2,000 labels" with equal seriousness, back the choice with an eval and a cost estimate, and know what to do when performance drifts
- * A track record of shipping – not just prototypes and papers, but systems that deliver value to real users
- * Experience working with scientific or scholarly text
- * Familiarity with AWS (S3, Batch, Lambda, SageMaker) and Parquet or Iceberg data lake patterns
- * Experience running LLMs under real cost and latency budgets in production
- * Some exposure to agentic AI applications: tool use, multi-step reasoning, guardrails, and evaluation of trajectories rather than single-turn outputs
Benefits
- Meeting-free Friday afternoons, allowing more time for heads-down work and professional development
- Continual learning and internal mobility
- A robust body of employee programming providing opportunities to foster community, learn, and grow
- A comprehensive benefits package
Company Overview
Company H1B Sponsorship