Edge AI Engineer: Data Pipelines & Clinical NLP

Tricorder Systems; United Kingdom; On-site; GBP 70,000 - 120,000; Full time

Benefits offered by this job

Equity

Flexible hours

Job summary

Tricorder Systems is seeking a hands-on engineer to build the evaluation harness and training data pipeline for AI models, and a clinical knowledge base grounded in SNOMED CT and ICD-10. The role spans data labelling, processing, and building a scalable knowledge base, with a focus on field deployment and edge cases.

You’ll own two workstreams alongside the founder, delivering a working eval harness and a first version of the knowledge base within the initial 3 months.

Qualifications

• Strong Python and data engineering experience.
• Experience building ML evaluation frameworks.
• Data labelling and training-data preparation experience.
• Working knowledge of databases and data modelling.
• Practical NLP experience: entity extraction/linking or document parsing.
• Ability to handle sensitive data under data protection and clinical safety constraints.
• Pragmatism and speed in a startup environment.

Responsibilities

• Own two workstreams: data pipeline/evaluation harness and clinical knowledge base.
• Design labelling schemas, tooling and guidelines with clinical input.
• Ingest, clean, de-identify, and transform multimodal data (video, sensor, text).
• Structure and version video datasets; ensure traceability and searchability.
• Prepare well-documented train/validation/test splits with lineage from raw data to model input.
• Build automated evaluation harness for vision-language and NLP models.
• Design the schema for the clinical knowledge base and map terminologies (SNOMED CT, ICD-10).
• Collaborate with clinicians to reflect real-world pre-hospital workflows.
• Deliver a working eval harness and initial knowledge base within 3 months.

Skills

Python

Data pipelines

NLP

SQL

Data labelling

Video data

De-identification

Tools

PostgreSQL

Airflow

Pandas

Job description

Tricorder Systems is seeking a hands-on engineer to build the evaluation harness and training data pipeline for AI models, and a clinical knowledge base grounded in SNOMED CT and ICD-10. The role spans data labelling, processing, and building a scalable knowledge base, with a focus on field deployment and edge cases.

You’ll own two workstreams alongside the founder, delivering a working eval harness and a first version of the knowledge base within the initial 3 months.