Senior Manager / Manager, LLM Architect, SIMFONI

Job description:

Overview
The Consortium for Clinical Research and Innovation Singapore (CRIS), a wholly owned subsidiary of MOH Holdings, was established in 2020 with the goal of strengthening synergies and promulgating strategies for national-level clinical research and translation programmes under the stewardship of the Singapore Ministry of Health.
CRIS brings together seven national R&D, clinical translation and service programmes to advance clinical research and innovation for Singapore, and establish important capabilities for a future-ready healthcare system.
The Business Entities under CRIS include:

• Singapore Clinical Research Institute (SCRI)
• National Health Innovation Centre (NHIC)
• Advanced Cell Therapy and Research Institute, Singapore (ACTRIS)
• Precision Health Research, Singapore (PRECISE)
• Singapore Translational Cancer Consortium (STCC)
• Cardiovascular Disease National Collaborative Enterprise (CADENCE)
• SIngapore Medical FOundation AI model (SIMFONI)

Together, CRIS makes a positive difference to Singapore patients and researchers by ensuring that these clinical research platforms and programmes are at the cutting edge of capability development and innovation. If you are as passionate as we are in clinical trials and research, we want you!
SIMFONI
The SIngapore Medical FOundation AI model (SIMFONI) Programme was established in 2025 to advance the safe and responsible use of Artificial Intelligence (AI) in Singapore’s public healthcare ecosystem to support healthcare professionals in providing care to patients.
SIMFONI will build up the capabilities and infrastructure to accelerate the development and deployment of large-scale machine learning models, known as Foundation models (FM) for improved healthcare outcomes.
About the Role
We are hiring a senior LLM/Foundation Model architect to provide technical leadership for a healthcare foundation model programme in Singapore. The programme focuses on developing, adapting, evaluating, and preparing large language models for healthcare use cases, including clinical decision support, healthcare knowledge assistance, and domain-specific model adaptation.
This role will work closely with multiple project teams and partners to review, advise, and support their LLM development plans. The candidate will provide expert guidance across model selection, continual pretraining, fine-tuning, evaluation design, infrastructure planning, deployment feasibility, and engineering risk management.
The successful candidate should be able to combine strong LLM technical depth with practical engineering judgment. They should be comfortable reviewing technical proposals, advising teams on implementation approaches, supporting technical problem-solving, challenging assumptions constructively, and helping teams improve the quality, feasibility, and robustness of their LLM development work.
This is a technical leadership role rather than a day-to-day coding role. However, the candidate should have sufficient hands-on engineering background and practical LLM development experience to provide credible, concrete, and actionable guidance to project teams.
Healthcare domain experience is highly desirable. Or candidates with strong industry experience in LLM or foundation model development, adaptation, evaluation, or deployment in other domains are also preferred.
Key Responsibilities
LLM Technical Leadership, Advisory, and Oversight
Serve as the programme-level technical lead for LLM architecture, model development strategy, training and tuning methodology, evaluation approach, and engineering feasibility.
Review, advise, and support project teams on technical proposals covering model selection, continual pretraining, fine-tuning, evaluation design, infrastructure assumptions, deployment feasibility, and risk mitigation.
Assess whether proposed methods are technically sound, feasible, scalable, reproducible, and aligned with programme objectives.
Provide practical technical guidance to help teams improve their model development plans, experiment design, engineering approach, and technical outcomes.
Establish review principles, reference approaches, and engineering guidelines for LLM development across the programme.
Support for Training, Tuning, and Evaluation Approaches
Advise teams on suitable approaches for continual pretraining, domain-adaptive pretraining, supervised fine-tuning, instruction tuning, preference tuning, RAG, grounding, tool use, and other LLM adaptation methods.
Review and support experiment design, baseline comparison, training stability, reproducibility, and model versioning, etc.
Assess evaluation strategies covering task performance, clinical relevance, safety, hallucination risk, robustness, calibration, bias, and model limitations.
Review model results and technical reports, and advise whether the evidence is sufficient for further development, pilot testing, or downstream deployment consideration.
Engineering and LLMOps Advisory
Advise project teams on engineering approaches for model development, experiment tracking, evaluation pipelines, model lifecycle management, model serving, monitoring, and continuous iteration.
Support teams in identifying scalable and maintainable engineering patterns, while working with dedicated implementation teams for actual delivery.
Work with infrastructure, platform, data, and solution architecture teams to ensure technical assumptions are realistic and aligned across workstreams.
Identify cross-cutting engineering risks related to scalability, reliability, observability, maintainability, cost, and operational feasibility.
Support the definition of technical review gates, model readiness criteria, and engineering quality expectations.
Cross-Team Technical Alignment
Facilitate technical design reviews, architecture discussions, model development reviews, and cross-team alignment sessions.
Translate clinical, research, and business objectives into technical review criteria and actionable gui