While the AI innovation race in Biopharma is focused on Drug discovery, Product Development/ CMC represents the next barrier/ bottleneck. The complexity of biological systems, the rigor of regulatory expectations, the pace of pipeline growth, and the enormous value at stake make this one of the highest-leverage domains for applied data science and AI in the entire pharmaceutical value chain.
We here at BTS - PDST, are building a dedicated, AI-native team that is driving cutting edge programs across early stage, late stage and commercial product development to accelerate E2E product development and launch, maximize yields of block buster products. Through our deep collaboration with PDST scientists we are boldly reimagining how Abbvie can bring our pipeline products and lifesaving drugs to patients faster, safer and in cost effective manner fueled by AI.
Principal Data Engineer is a highly technical, AI-native role responsible for designing, building, and operating production-grade data pipelines and data products that power AI/ML, analytics, and automation across AbbVie's CMC and manufacturing ecosystem.
This role is embedded inside PDST and works at the frontier of pharmaceutical data engineering. You will integrate and harmonize data from the full spectrum of manufacturing and development systems — including MES, historians, LIMS, QMS, ERP, and instrument platforms — and transform it into reliable, governed, semantically rich data assets that data scientists, process engineers, and AI systems can actually use.
- Enterprise-scale scope: Enterprise-scale biologics portfolio spanning clinical, commercial, and lifecycle stages
- Building AI playbook for the future: First-in-AbbVie and first-in-biologics analytical approaches; you build the AI playbook for the future
- Growth and Impact: Direct impact on regulatory submissions, commercial readiness, and manufacturing decisions through deep cross-functional exposure to manufacturing, quality, regulatory, and scientific leadership
- Mission: Every model you build helps ensure safe, reliable medicines reach patients at scale
Responsibilities
Data Ingestion & Integration
- Design and implement scalable, robust data ingestion pipelines that connect CMC and manufacturing source systems — including MES (Manufacturing Execution Systems), process historians, LIMS, QMS, ERP platforms, and instrument data sources — to centralized and federated data environments.
- Build connectors, adapters, and integration layers that handle the heterogeneous data formats, protocols, and latency profiles characteristic of pharmaceutical manufacturing environments.
- Support both batch and real-time/streaming data patterns, selecting appropriate architectures based on use case requirements.
Data Harmonization & Semantic Modeling
- Develop and maintain harmonized data models and ontologies that bring consistency to CMC and manufacturing data across sites, systems, and modalities.
- Execute semantic mapping efforts that align source system fields, units, and identifiers to enterprise data standards and scientific meaning.
- Collaborate with process scientists, analytical chemists, and manufacturing engineers to ensure data models accurately reflect domain reality.
Data Quality, Observability & Governance
- Implement automated data quality controls, validation frameworks, and anomaly detection mechanisms across pipeline layers.
- Build and maintain data lineage documentation and metadata infrastructure, enabling full traceability from source system to AI model input.
- Establish pipeline observability practices — monitoring, alerting, SLA tracking — to ensure data product reliability in production.
- Support data governance practices aligned with GxP requirements, 21 CFR Part 11, and AbbVie data standards.
AI/ML Enablement & Data Product Development
- Architect and deliver governed, versioned, reusable data products purpose-built for AI/ML consumption, including feature stores, curated datasets, and vector-ready data layers for RAG and LLM applications.
- Partner closely with data scientists, ML engineers, and process modelers to understand model data requirements and translate them into reliable, scalable data infrastructure.
- Accelerate AI program delivery by eliminating data bottlenecks — not by workarounds, but by solving root causes structurally.
Platform & Operational Enablement
- Contribute to the design and evolution of PDST's cloud-based data platform, including lakehouse architecture, data cataloging, access control, and compute infrastructure.
- Write and maintain infrastructure-as-code, CI/CD pipelines, and automated testing frameworks for data systems.
- Support platform onboarding of new CMC data domains and manufacturing sites, ensuring consistent application of standards and patterns.
- Provide operational support for production data pipelines, maintaining uptime and data freshness commitments.
Stakeholder Engagement & Scientific Leadership
- Influence technical decision-making without formal authority — earning trust through scientific rigor, transparent methodology, and demonstrated business impact.