Dyad is seeking an NLP Engineer to join our Applied AI team and work on the clinical document understanding pipeline that underpins BetterLetter and related products.
This is a hands‑on engineering role focused on building, improving, and maintaining production NLP systems. You will work on OCR‑aware document processing, entity extraction and linking, and the safe integration of LLM components within a constrained, regulated architecture.
The role is offered on a hybrid basis from our London office.
Core responsibilities
- Design, build, and maintain NLP pipelines for clinical document processing using Python.
- Develop and extend pipeline components as well as training configurations, packaging, and versioning. Refactor and improve pipeline components for maintainability, scalability, and clarity.
- Train, evaluate, and deploy NLP and OCR models for clinical concepts. Maintain evaluation datasets and implement regression testing for model and pipeline updates.
- Improve document structure detection, sectioning, and layout‑aware extraction, particularly for scanned documents.
- Enhance handling of negation, temporality, and related concepts in clinical text.
- Analyse production errors and implement targeted improvements to reduce recurring extraction and coding issues.
- Integrate LLM‑based components into the pipeline using structured inputs and validated outputs. This includes implementing schema validation, rule‑based checks, and other guardrails around model outputs.
- Optimise pipeline performance, including latency, throughput, and cost per document.
- Collaborate with Engineering to support production deployment and monitoring of NLP components.
Requirements
- Minimum of a bachelor's degree in computer science, computational linguistics, or equivalent educational attainment.
- At least 2 years of commercial experience. This is not a graduate role.
- Strong professional experience in applied NLP and machine learning engineering.
- Advanced Python skills, including experience building and maintaining production ML systems.
- Hands‑on experience with common NLP frameworks.
- Experience training and evaluating NER and/or entity‑linking models.
- Experience working with noisy or unstructured text data, such as OCR‑derived documents.
- Familiarity with combining rule‑based and statistical approaches in production systems.
- Experience designing and implementing evaluation metrics and benchmarks as well as regression testing for NLP systems.
- Experience working with healthcare or clinical text.
- Familiarity with clinical terminologies such as SNOMED CT.
- Experience integrating LLMs into structured application pipelines.
- Experience working in regulated or high‑assurance environments.
- Exposure to hybrid symbolic and generative AI architectures.
Personal attributes
- Detail‑oriented with a strong focus on accuracy and reliability.
- Pragmatic approach to problem‑solving, selecting appropriate techniques for the task.
- Comfortable working in a fast‑paced startup environment.
- Strong communication skills and ability to work effectively within a multidisciplinary team.
Benefits
- Company pension.
- 25 days of paid annual leave (pro‑rata).
- Flexible hybrid working environment.
- Employee Assistance Programme.
- Modern, dog‑friendly office near Chancery Lane with free drinks.
#J-18808-Ljbffr