Key Responsibilities
Strategy & Governance
- Define and execute the AI/ML roadmap aligned to clinical PV and quality/audit objectives (e.g. site selection patient recruitment signal detection case intake automation RBQM).
- Run an AI opportunities and triage process (usecase assessment value/risk scoring resourcing and portfolio tracking).
- Establish AI governance: model risk classification validation/qualification plans algorithm change control documentation (model cards datasheets) and monitoring.
- Ensure compliance with GxP 21 CFR Part 11/Annex 11 ICH E6(R3) GVP for PV data privacy (GDPR/PHI) and vendor and Ergomed IT and security standards.
- Define and track KPIs (e.g. model adoption cycle-time reduction accuracy/recall signal sensitivity/specificity audit findings cost-to-serve).
Solution Design & Delivery
- Understands and translates business and functional needs into AI and machine learning problem statements
- Translates complex AI and machine learning problem statements into specific deliverables and requirements
- Designs and develops scalable solutions that leverage machine learning and deep learning models to meet Ergomed business requirements
- Performs tasks such as data normalization feature selection dimensionality reduction and handling missing or outlier data to ensure data quality and relevance
- Works closely with business process analysts and data engineers to frame business problems and need into AI/ML product or workflow. Collaborates with development teams to test and deploy machine learning models
- Evaluate models performance conduct cross-validation and use metrics such as accuracy precision recall or F1-score to assess model quality
- Monitor the performance of deployed models track data or concept drift and update or retrain models as needed
- Stays updated with the latest advancements in AI and ML technologies to identify their potential applications in the Ergomed business model and internal processes improvements
Data Standards Quality & Interoperability
- Knowledge and understanding of use of CDISC SDTM/ADaM HL7 FHIR MedDRA WHODrug and SNOMED CT.
- Establish robust data quality frameworks (profiling completeness timeliness conformance checks) and FAIR data principles.
- Oversee PHI/PII handling de-identification and secure data access patterns (role-based access least privilege).
Execution Coordination & Stakeholder Management
- Educate mentor and grow more junior data scientists ML engineers and AI specialists; cultivate a culture of scientific rigor design-for-compliance and continuous learning.
- Partner with Clinical Ops PV/Drug Safety Biostatistics Regulatory QA IT/Security and commercial client teams to translate needs into high-value AI solutions.
- Collaborate and cooperate with external partners and vendors and ensure SOWs SLAs and validation deliverables meet compliance and performance expectations.
- Oversee design development validation and deployment of ML/NLP/GenAI solutions for:
- Clinical Operations: protocol feasibility site and investigator selection patient eligibility/matching country start-up forecasting RBQM/central monitoring query reduction.
- Pharmacovigilance: ICSR intake/triage (NLP) de-duplication case validity checks literature screening medical coding support signal detection & prioritization causality/seriousness assistance (human-in-the-loop).
- Regulatory & Quality: authoring support for submissions and SOPs (GenAI with RAG) audit analytics inspection readiness dashboards document classification in eTMF/CTMS.
- RWD/RWE & Safety Analytics: EHR/claims ingestion de-identification pseudonymization cohort building outcomes and safety signal analytics.
- Ensure solutions integrate with existing platforms (e.g. EDC CTMS eTMF Safety DBs like Argus/LifeSphere Veeva stacks Ergomed analytical data platform and BI tool).
- Drive MLOps in a regulated setting: CI/CD for models feature stores reproducible pipelines lineage/metadata monitoring (data drift performance) and rollback.
- Communicate complex AI concepts and model risk/impact to executive and non-technical stakeholders with clarity and accountability.
Qualifications :
Job Requirements
Education
- Bachelors degree in computer science data science engineering biostatistics or related field (masters or Ph.D. preferred).
- Desirable: coursework or certification in GxP/CSV biostatistics clinical trials pharmacovigilance or regulatory affairs.
Experience
- 710 years in applied AI/ML (including NLP/GenAI) with 24 years coordinating cross-functional teams or programs.
- Demonstrated delivery of production AI products within business processes.
- Hands-on leadership of MLOps and lifecycle management in cloud (AWS GCP Azure); experience with Azure ML Databricks Synapse/Fabric MLflow
- Experience integrating with life sciences systems (e.g. EDC: Medidata/Oracle; CTMS/eTMF: Veeva; Safety: Argus/LifeSphere; BI: Power BI).
- Experience in creating agentic multi-agent AI solutions.
Preferred/Bonus
- Experience with RWD/RWE de-identification privacy engineering.
- Knowledge of ICH E6(R3) GVP Modules MedDRA/WHODrug coding and CDISC implementation.
- Familiarity with EU AI Act considerations for high-risk systems and internal Model Risk Management frameworks.
Skills
- Technical:
- Python SQL; ML frameworks (scikit-learn XGBoost PyTorch/TensorFlow); NLP/LLMs (Transformers RAG prompt engineering evaluation).
- Data engineering (Spark) pipelines/orchestration (e.g. Azure Data Factory DataBricks Fabric LakeHouse Airflow) feature stores containers (Docker/Kubernetes).
- MLOps (MLflow/Azure ML) experiment tracking monitoring CI/CD testing and observability.
- BI & analytics (e.g. Power BI Tableau QlikSense) statistical methods (sampling hypothesis testing time series survival/longitudinal analysis a plus).
- Interoperability & standards (CDISC FHIR MedDRA WHODrug); data privacy (GDPR/PHI) and de-identification.
- Compliance & Quality:
- Computerized System Validation (CSV) documentation discipline traceability audit trail management change control SOP authorship.
- Bias/explainability tools human-in-the-loop controls risk assessments and control design for regulated AI.
- Leadership & Communication:
- Team building and mentoring; stakeholder engagement across Clinical Ops PV QA and Regulatory.
- Clear communication of risk/benefit assumptions limitations and model performance to non-technical audiences.
- Product mindsettranslating user needs into validated usable solutions with measurable outcomes.
Additional Information :
We prioritize diversity equity and inclusion by creating an equal opportunities workplace and a human-centric environment where people of all cultural backgrounds genders and ages can contribute and grow.
To succeed we must work together with a human first approach. Why because our people are our greatest strength leading to our continued success on improving the lives of those around us.
We offer:
Training and career development opportunities internally
Strong emphasis on personal and professional growth
Friendly supportive working environment
Opportunity to work with colleagues based all over the world with English as the company language
Our core values are key to how we operate and if you feel they resonate with you then Ergomed is a great company to join!
Quality
Integrity & Trust
Drive & Passion
Agility & Responsiveness
Belonging
Collaborative Partnerships
We look forward to welcoming your application.
#LI remote
Remote Work :
No
Employment Type :
Full-time