Postdoctoral Research Associate in Clinical NLP and MLOps
Drive clinical NLP and small language model optimisation for the Horizon Europe AIM-SAFE healthcare project.
Best for:Best for postdoctoral researchers with expertise in NLP, MLOps, and parameter-efficient model fine-tuning.
Overview
The University of Liverpool is seeking a Postdoctoral Research Associate to take technical ownership of Work Package 5, Task 5.1 within the Horizon Europe AIM-SAFE project. The researcher will benchmark and select open-weight Small Language Models for clinical tasks, design MLOps fine-tuning pipelines, and implement quantisation for efficient NHS deployment. Responsibilities also include collaborating with international consortium partners such as CERN and KAIST, publishing high-impact research, and mentoring junior researchers. This fixed-term role runs until May 2030 within the Intelligent Systems Research Group.
What you'll do
- Benchmark and select open-weight Small Language Models for clinical NLP tasks including entity recognition and drug-interaction reasoning
- Design and implement parameter-efficient fine-tuning pipelines such as LoRA, QLoRA, and adapters within a reproducible MLOps framework
- Apply quantisation techniques including GPTQ, AWQ, and GGUF/llama.cpp to enable CPU-feasible deployment in clinical settings
- Integrate federated and privacy-preserving learning mechanisms via the CAFEIN platform in collaboration with CERN
- Embed clinical safety governance layers and lead the delivery of project Deliverable D5.1
- Act as technical liaison with multi-agent orchestration and retrieval-augmented generation teams across partner institutions
- Publish research findings in high-impact academic venues and mentor junior researchers within the research group
Requirements
Degree required: PhD in Computer Science or a closely related discipline (or near completion)
Language: English; professional working proficiency
Instruction: English
Publications: Evidence of high-quality research output commensurate with career stage
- Demonstrated experience in natural language processing and machine learning
- Proficiency in fine-tuning open-weight language models using parameter-efficient methods
- Experience with MLOps frameworks and reproducible model training pipelines
- Familiarity with model quantisation methods for resource-constrained environments
- Strong track record of academic publishing in relevant peer-reviewed venues
- Must hold a PhD or be awaiting award upon appointment
Funding & terms
How to apply
Deadline: 2026-11-12
Candidates must submit their application online through the university's recruitment portal before the closing date. Ensure all required personal and professional details are provided in the application form.
- CV
Create a free profile to unlock your personal match score, the full requirements, funding details and how to apply — in English or Persian.