AdvanSix · 4 weeks ago
Manager, Unified Data Platform & AI Engineering
AdvanSix plays a critical role in global supply chains, innovating and delivering essential products for various end markets. The Manager of Unified Data Platform & AI Engineering will own the design, build, and run of the digital data platform, managing a team to deliver a governed Unified Data Layer and robust data products.
ChemicalManufacturing
Responsibilities
Own the UDL reference architecture
Define and operate secure OT/IT integration patterns
Partner with SAP teams to integrate S/4HANA and SAP DataSphere into the semantic layer strategy
Lead engineering for ingestion (batch/CDC/streaming), transformation (SQL/PySpark), testing, observability, and SLAs
Implement reusable data product patterns: naming conventions, contracts, quality rules, SCD handling, and documentation “readmes.”
Expose governed access for Power BI, APIs, and Copilot/agents; prevent direct system scraping
Stand up ML platform services: feature store, model registry, experiment tracking, inference endpoints, monitoring (drift, accuracy, latency), and A/B controls
Partner with Data Science to productionize models with SLOs and rollback runbooks
Establish CI/CD for data and ML (branching, environments, approvals, blue/green)
Implement observability: pipeline health, freshness, DQ, lineage coverage, cost dashboards, and alerting/On-Call
Enforce RBAC, secrets management, PII/HSE classifications, data retention, and DLP alignment with the Power Platform CoE
Manage a 3–5 FTE engineering team; recruit, coach, and develop talent
Direct vendor pods for ingestion, historian connectors, DataSphere modeling, and BI enablement; define DoD, SLAs, and knowledge transfer to insource sustainably
Collaborate with Data Governance/MDM, Reporting & BI, Automation, Cyber/IT, and plant controls teams
Embed data quality rules, lineage, KPI canon, and MDM keys/survivorship into pipelines
Champion documentation, runbooks, post-incident reviews, and change control—especially for OT integrations
Qualification
Required
Minimum 7 years' in data/platform engineering with 3+ years leading teams or tech leads
Proven delivery of a Lakehouse or equivalent enterprise data platform and streaming/CDC pipelines
Strong hands-on with Azure data stack; or equivalent in AWS/GCP with ability to translate to Azure
Deep SQL and one major programming language (Python preferred); PySpark experience for large-scale transforms
Practical MLOps/DevOps: Git-based CI/CD, IaC, model registry/monitoring, containerized services, observability
Understanding of OT systems and historians, DCS/PLC data, OPC UA/MQTT patterns, and OT cybersecurity considerations (ISA-95/99)
Working knowledge of SAP S/4HANA data structures and SAP DataSphere semantic modeling; ability to partner with SAP teams
Familiarity with Power BI consumption patterns (semantic models, RLS) and integration with certified datasets/APIs
Manufacturing or process industry experience strongly preferred
Benefits
Paid holidays
Paid time off including vacation
Eligibility to purchase company stock
Tuition reimbursement
401K with a competitive company match
Medical, dental and vision insurance
Flexible spending and health savings account eligibility
Employer-provided short term disability benefits
Eligibility to purchase long term disability benefits
Employer-provided basic life insurance
Eligibility to purchase voluntary life coverages
Company
AdvanSix
AdvanSix is a chemical industry
H1B Sponsorship
AdvanSix has a track record of offering H1B sponsorships. Please note that this does not
guarantee sponsorship for this specific role. Below presents additional info for your
reference. (Data Powered by US Department of Labor)
Distribution of Different Job Fields Receiving Sponsorship
Represents job field similar to this job
Trends of Total Sponsorships
2023 (2)
2022 (1)
2021 (1)
2020 (1)
Funding
Current Stage
Late StageTotal Funding
$12MKey Investors
U.S. Department of Agriculture
2024-09-16Grant· $12M
Recent News
2025-12-24
2025-12-07
2025-11-08
Company data provided by crunchbase