AI Product Owner (ML & GenAI) for Operations

Descripción del puesto:

About the job

Mission of the role:
Own AI products for Operations - end to end.
* Act as Product Owner for AI projects and products in Operations: vision, roadmap, backlog, and delivery with factories, functions, zones, IT, OT, and external partners.
* Understand the mechanics behind the product (data, models, architecture, workflow) - and the process around it.
* Be hands-on: work with data, build and evaluate models, and prototype solutions, not only coordinate others.
* Take use cases from ideation to deployment: change management, user coaching, and value tracking against the business case.
* Deliver PoCs, scale what works, kill what does not, and build in-house ML/GenAI capability.
* Define and run sustainable AI development and management practices in line with Responsible AI.
Product ownership
* Own the product roadmap for assigned AI products (ML platforms and GenAI solutions) and keep it aligned with Operations priorities and the business case.
* Translate business problems into use cases, data needs, model choices, and a sequenced backlog.
* Run the product with cross-functional teams: process / IWS, OT, data, IT, factories, and partners. Decide what is in, what is out, and when it is good enough to go live.
* Set acceptance criteria, go-live criteria, and model-evaluation standards. Challenge black-box delivery until mechanics and quality are understood.
Hands-on ML / data / GenAI
* Work with industrial data: tags, historians, time series, data gaps, cleansing, feature engineering.
* Build, train, evaluate, and iterate models (anomaly detection, predictive / remaining-useful-life style use cases, classical ML, and GenAI).
* Ensure a consistent, state-of-the-art architecture across products; experiment with new ML/GenAI techniques and bring them into the stack (Azure / Databricks and related tools).
* Deliver in-house solutions and integrations; test, troubleshoot, and debug until they work in a plant context.
Users, change, value
* Work with users from ideation to deployment: workshops, shadowing on the line, demos, training, hypercare.
* Drive change management so alerts and tools are used - not only installed. Coach key users and application owners.
* Track value (e.g. unplanned downtime, PR, quality, cost) versus the business case; feed results back into roadmap and model priorities.
* Document learnings and best practices so successful products can be scaled to further sites/lines.
Lab / CoE
* Animate the centre of excellence and AI self-service where it helps adoption.
* Mentor others and grow in-house capability so Operations is not dependent on a single expert or vendor.
Indicative KPIs
* Roadmap delivery: committed site/line onboardings and product increments on time.
* Model / product quality: agreed evaluation criteria met before go-live; alert precision/usefulness in operation.
* Adoption: share of modelled assets / target users actually using the product (actioned alerts, active companions).
* Value: realized vs business-case impact , with a clear measurement method.
* Lab throughput: PoCs taken to a working product or explicitly stopped, with documented learnings (indicative ~4-5 prototypes/year, 2-3 months each).
Skills needed for the role
* Product ownership in a technical domain: roadmap, backlog, prioritization, stakeholder management, and the ability to say no. Comfortable as the single owner of an AI product, not as a pure project coordinator.
* Solid ML foundation: supervised/unsupervised learning, time-series and anomaly detection, feature engineering, model evaluation, and the judgement to choose the right model for the loss - not the most fashionable one.
* Hands-on with data: exploratory analysis, industrial/OT data (sensors, tags, historians), data quality, and working with process engineers on loss trees and critical equipment.
* GenAI literacy: foundation models, orchestration (e.g. LangChain / similar), and when GenAI is the right tool vs classical ML.
* Proficiency in at least one language used for data/ML work (Python preferred; Java/C# a plus). Comfortable writing code, not only slides.
* Experience with cloud data/ML platforms (preferably Azure, Databricks).
* Change management and user empathy: can work with operators, maintainers, supply chain, procurement and leadership from idea through go-live.
* Value tracking: can turn a use case into a measurable business case and keep score after deployment.
* Strong collaboration and communication: explain models and trade-offs to non-technical stakeholders; work with IT, Data, OT, and partners.
* Innovative but pragmatic: experiment fast, stay current, ship what plants will use.
* Fluent English.

* Bachelor’s or Master’s in Computer Science, Engineering, Data Science, Industrial Engineering, or a related field. PhD is a plus; equivalent hands-on experience equally valued.
* Professional experience in IT/Data/AI, with a mix of machine learning delivery and product ownership (or a technical lead who has owned a product end to end).
* Demonstrated experience taking an ML or AI solution from idea to users in a live operational environment — including messy data, model iteration, and adoption — not only a notebook or a vendor demo.
* Experience with industrial or operations context is a strong plus (manufacturing, maintenance, process, OT/IT).
* Experience working with cross-functional teams, plants/sites, and external partners.
* Experience with cloud platforms (Azure, AWS, or GCP) and with programming/frameworks used in ML and backend development.
* Project/product delivery track record: multiple workstreams, clear priorities, on-time delivery." descriptionHeader="
Mission of the role:
Own AI products for Operations — end to end.
* Act as Product Owner for AI projects and products in Operations: vision, roadmap, backlog, and delivery with factories, functions, zones, IT, OT, and external partners.
* Understand the mechanics behind the product (data, models, architecture, workflow) — and the process around it.
* Be hands-on: work with data, build and evaluate models, and prototype solutions, not only coordinate others.
* Take use cases from ideation to deployment: change management, user coaching, and value tracking against the business case.
* Deliver PoCs, scale what works, kill what does not, and build in-house ML/GenAI capability.
* Define and run sustainable AI development and management practices in line with Responsible AI.
Product ownership
* Own the product roadmap for assigned AI products (ML platforms and GenAI solutions) and keep it aligned with Operations priorities and the business case.
* Translate business problems into use cases, data needs, model choices, and a sequenced backlog.
* Run the product with cross-functional teams: process / IWS, OT, data, IT, factories, and partners. Decide what is in, what is out, and when it is good enough to go live.
* Set acceptance criteria, go-live criteria, and model-evaluation standards. Challenge black-box delivery until mechanics and quality are understood.
Hands-on ML / data / GenAI
* Work with industrial data: tags, historians, time series, data gaps, cleansing, feature engineering.
* Build, train, evaluate, and iterate models (anomaly detection, predictive / remaining-useful-life style use cases, classical ML, and GenAI).
* Ensure a consistent, state-of-the-art architecture across products; experiment with new ML/GenAI techniques and bring them into the stack (Azure / Databricks and related tools).
* Deliver in-house solutions and integrations; test, troubleshoot, and debug until they work in a plant context.
Users, change, value
* Work with users from ideation to deployment: workshops, shadowing on the line, demos, training, hypercare.
* Drive change management so alerts and tools are used — not only installed. Coach key users and application owners.
* Track value (e.g. unplanned downtime, PR, quality, cost) versus the business case; feed results back into roadmap and model priorities.
* Document learnings and best practices so successful products can be scaled to further sites/lines.
Lab / CoE
* Animate the centre of excellence and AI self-service where it helps adoption.
* Mentor others and grow in-house capability so Operations is not dependent on a single expert or vendor.
Indicative KPIs
* Roadmap delivery: committed site/line onboardings and product increments on time.
* Model / product quality: agreed evaluation criteria met before go-live; alert precision/usefulness in operation.
* Adoption: share of modelled assets / target users actually using the product (actioned alerts, active companions).
* Value: realized vs business-case impact , with a clear measurement method.
* Lab throughput: PoCs taken to a working product or explicitly stopped, with documented learnings (indicative ~4–5 prototypes/year, 2–3 months each).
Skills needed for the role
* Product ownership in a technical domain: roadmap, backlog, prioritization, stakeholder management, and the ability to say no. Comfortable as the single owner of an AI product, not as a pure project coordinator.
* Solid ML foundation: supervised/unsupervised learning, time-series and anomaly detection, feature engineering, model evaluation, and the judgement to choose the right model for the loss — not the most fashionable one.
* Hands-on with data: exploratory analysis, industrial/OT data (sensors, tags, historians), data quality, and working with process engineers on loss trees and critical equipment.
* GenAI literacy: foundation models, orchestration (e.g. LangChain / similar), and when GenAI is the right tool vs classical ML.
* Proficiency in at least one language used for data/ML work (Python preferred; Java/C# a plus). Comfortable writing code, not only slides.
* Experience with cloud data/ML platforms (preferably Azure, Databricks).
* Change management and user empathy: can work with operators, maintainers, supply chain, procurement and leadership from idea through go-live.
* Value tracking: can turn a use case into a measurable business case and keep score after deployment.
* Strong collaboration and communication: explain models and trade-offs to non-technical stakeholders; work with IT, Data, OT, and partners.
* Innovative but pragmatic: experiment fast, stay current, ship what plants will use.
* Fluent English." workFromHome="Hybrid