NLP & Scientific Reasoning Post-Doc

Position Overview

NGRF is seeking a researcher to join the LLM Lab and contribute to NGRF's scientific discovery platform. This role combines original research in scientific integrity with hands-on development of NLP and AI capabilities on a large-scale biomedical data platform.

Key Responsibilities

• Design and execute large-scale computational analyses of scientific publishing practices across the biomedical literature.
• Publish findings in peer-reviewed venues (e.g., Scientometrics, JASIST, Quantitative Science Studies).
• Develop and validate models for semantic analysis of scientific text — including claim extraction, relevance scoring, and relationship detection.
• Build on existing RAG infrastructure and vector search capabilities.
• Work with a multi-terabyte literature database and external APIs (PubMed, CrossRef, OpenAlex) to build scalable data processing and analysis pipelines.
• Integrate structured and unstructured data sources into reproducible workflows.
• Collaborate with external partners in the research integrity community.
• Present findings at conferences and contribute to NGRF's visibility in the scientific integrity space.

Required Qualifications

• PhD in natural language processing, computational linguistics, computer science, information science, or a related field.
• Publication record in NLP, text mining, or computational approaches to scientific literature.
• Strong programming skills in Python with experience in modern ML/NLP frameworks (e.g., PyTorch, Hugging Face Transformers).
• Experience working with large text corpora and large-scale datasets.
• Familiarity with vector databases, embeddings, semantic search, and retrieval-augmented generation (RAG).
• Experience with knowledge graphs, ontology design, or graph databases.
• Demonstrated ability to design and execute independent research projects.
• Strong written and oral communication skills.

Preferred Qualifications

• Background in scientometrics, bibliometrics, or research integrity.
• Experience with scientific publishing APIs and metadata (DOIs, PubMed, CrossRef, OpenAlex).
• Familiarity with claim extraction, relation extraction, or scientific argument mining.
• Proficiency with Git, GitHub, and collaborative software development workflows.
• Experience with Linux/Unix environments and command-line tools.
• Familiarity with containerization (Docker) and CI/CD pipelines.
• Experience with AI-assisted development tools (e.g., Claude Code, GitHub Copilot).
• Experience deploying NLP models in production or near-production settings.
• Familiarity with Kubernetes, infrastructure-as-code, or cloud platforms (AWS).
• Experience with SQL databases (PostgreSQL) and data modeling.