As part of the AITP organisation, the Sight Vertical is responsible for building an end-to-end business around AI/ML anomaly detection products, operating as two product families.
Sight.Sense detects anomalies in parametric data: Sense.Machine (MiniTerms) for machine performance anomaly prediction and optimisation; Sense.Tooling (AI ToolSense) for tooling life optimisation; Sense.Energy for optimisation of supplies consumption; Sense.Stamping for stamping process anomaly prediction and surface splits; and Sense.MILO for in-process test quality anomaly detection.
Sight.Vision detects anomalies in image data: Vision.Scout for in-station visual error proofing across manual and automated processes; Vision.Surface for surface defect detection in components and vehicle bodies through the complete manufacturing process; and Vision.Station for manual process monitoring and anomaly detection covering productivity, quality and ergonomics.
The Sight Vertical product portfolio are set to converge on the Sight Anomaly Platform, which builds on Ford's Core Systems layer (Core OS, CoreDS, Core AI and CoreApps/FPS+). Together they form Ford's scalable platform to deliver, process quality assurance, and zero-downtime, zero-defect manufacturing.
How the Sight Vertical Works
The Sight Vertical operates through value-focused, cross-functional product teams across two product families: Sight.Sense and Sight.Vision, each led by a Head of Product Family. Product Managers lead the product problem, desired outcome and priority. Engineers and product designers work with them to discover and deliver effective solutions. Engineering Managers retain people, capacity, technical-system and engineering-health accountability.
This is a vertical-level role working across both product families. Priorities are agreed between the two Heads of Product and their Engineering Managers, so that your effort is directed by portfolio value. In addition, ATP Core provides the non-product engineering expertise the verticals depend on — enterprise standards, architecture, industrial controls, cybersecurity and platform guidance. This role holds a matrix reporting line into ATP Core to ensure alignment with those standards and ready access to that expertise.
The Opportunity
We are looking for a Data Engineer to build and operate the data foundations of the Sight portfolio — the pipelines that carry factory truth from PLCs, sensors, cameras and plant systems into a platform where anomalies can be detected reliably, at scale, across Ford's global manufacturing footprint.
Sight data is genuinely hard: tens of thousands of monitored assets, billions of waveforms annually, high-volume image and video data from Vision.Scout, Vision.Surface and Vision.Station, event-driven and polled sources, mixed plant infrastructure, and legacy equipment that was never designed to be observed. Getting this right is what makes the models trustworthy — and model trust is what determines whether plants act on our products or ignore them.
You will work at the intersection of operational technology, cloud data engineering and machine learning enablement, across both parametric data and image-based detection.
This is a senior individual contributor role. As the senior data engineering voice in the Sight Vertical, you will set the data engineering standards others work to, mentor engineers working with Sight data, and be accountable for ensuring the data foundations do not depend on you personally to remain trustworthy.
The Position
You will report into the Sight Vertical Director, with a matrix reporting line into the ATP Core team. The Sight Vertical Director leads what you prioritise and holds performance and development accountability; ATP Core provides technical alignment, enterprise standards and access to non-product expertise.
Working embedded within product streams across both families, you will partner with the Software Architect, Controls Engineer, Senior Software Engineer, Software Engineers, data scientists, the Launch and Deployment Engineers and plant controls teams. You will provide technical direction and mentorship to engineers on data-related work.
This is a senior individual contributor position with indirect leadership expectations.
You will work primarily on Google Cloud Platform, Ford's strategic cloud platform, and engage with CoreDS and Core AI as the strategic data and ML platform direction for the Sight Vertical.
Requires availability to travel internationally on occasion to understand plant data sources at source.
4 Days On Site In GTBC FORD México (Naucalpan, México)
Able to travel periodically to Europe and US.
Build and Operate Data Pipelines
Design, build, test and operate reliable ingestion and processing pipelines from plant OT sources — PLC event data, sensor streams, waveform captures, energy meters and image capture — into GCP.
Work with both event-driven ('notifier') and polled ('questioner') acquisition models, adapting to plant-level infrastructure choices such as CrossPLC (Ford developed middleware), Node-RED, DeviceWise, HiveMQ.
Build batch and streaming pipelines on GCP — Pub/Sub, Dataflow, Cloud Run/Functions, Cloud Composer, BigQuery and Cloud SQL/PostgreSQL — with infrastructure and pipelines defined as code.
Handle high-volume unstructured image and video data for Vision products, including object storage design, lifecycle and cost management, frame sampling and metadata indexing.
Own pipelines through production operation: monitoring, alerting, cost, backfill, recovery and incident response.
Data Modelling, Quality and Semantics
Design and maintain data models and semantic layers that make anomaly detection data usable across products and plants — including the common data model that enables global interoperability and rapid replication.
Implement data quality controls: validation, completeness, drift, latency and lineage — and make data health visible rather than assumed.
Model plant, line, station, asset and tool hierarchies consistently across both product families, integrating with systems of record such as Maximo.
Optimise storage, partitioning and query performance and cost in BigQuery.
Define and design data set lifecycle, historical, archiving, and purging approaches.
Enabling Analytics and Machine Learning
Prepare and serve training and inference datasets for parametric and image-based anomaly detection models, including labelling and annotation workflows, dataset versioning and feature availability.
Partner with data scientists on the MLOps path to production — reproducible datasets, versioning, monitoring and retraining triggers.
Support edge inference requirements, ensuring data available on the machine matches what models were trained on.
Build the datasets behind value evidence and dashboards used by plants, Product Managers and the ATP ledger.
Governance, Security and Engineering Practice
Apply Ford data governance, privacy, retention and cybersecurity requirements across OT and IT boundaries, working with the Controls Engineer where plant-floor constraints apply.
Give particular attention to imagery of people, where Vision.Station monitors manual processes and ergonomics — applying privacy, consent, retention and works council requirements with HR, Legal and local plant leadership.
Manage access and identity appropriately for plant, product and enterprise consumers of Sight data.
Define and own reusable ingestion and onboarding patterns so adding a plant is configuration rather than a project.
Set data engineering standards for the vertical — pipeline design, testing, documentation, naming, lineage and runbook expectations — and hold work to them through review.
Mentor and coach engineers working with Sight data, pairing on complex pipeline and modelling problems and explaining trade-offs rather than simply supplying answers.
Actively reduce single-person dependency on critical datasets and pipelines, ensuring at least one other engineer can maintain each of them.
Lead technical design for significant data platform changes, producing options, trade-offs and architecture decision records with the Software Architect.
Functional & Technical Knowledge
Minimum Requirements
Bachelor's degree in Computer Science, Data Engineering, Software Engineering or a related technical field, or equivalent demonstrable experience.
Typically 5+ years of hands-on data engineering experience, with a substantial track record of production pipelines and data platforms you have designed, built and operated at scale.
Advanced SQL and strong Python, with the judgement to choose appropriate tools and patterns rather than defaulting to the familiar.
Practical experience with a major cloud data platform — GCP required (BigQuery, Pub/Sub, Dataflow, Cloud SQL, Cloud Storage, Composer).
Experience with batch and streaming data processing and with orchestration tooling.
Experience of data modelling for analytical and operational use, and of data quality and validation practice.
Software engineering fundamentals: version control, automated testing, CI/CD (GitHub Actions), infrastructure as code.
Ability to work with imperfect, real-world industrial data and to trace problems back to source.
Demonstrable experience mentoring and growing other engineers — pairing, design and code review, technical coaching and knowledge transfer.
Ability to lead complex technical work end to end under ambiguity, and to communicate data trade-offs clearly to product, plant and non-technical stakeholders.
Fluent professional English.
Team player, problem solver, agile way of working.
Preferred Requirements
Experience with time-series, high-frequency sensor or waveform data at scale.
Experience with unstructured image or video data at scale, including annotation pipelines and dataset versioning for computer vision models.
Experience with industrial and OT data sources — PLC tag data, OPC-UA, MQTT, Modbus, historians, SCADA, energy metering.
Experience supporting ML workloads and MLOps (Vertex AI, Keras/TensorFlow, feature stores, model monitoring).
Experience with dbt, Dataform, Terraform or equivalent transformation and infrastructure-as-code tooling.
Familiarity with data governance and cataloguing in a large enterprise, and with OT/IT security segregation.
Experience of data privacy requirements where imagery or video of people is processed.
Experience acting as the senior or lead data engineer in a team, setting standards and owning a data platform's direction.
Exposure to manufacturing systems (Maximo, MES, GDPS 2.0) or automotive and other high-complexity manufacturing.
Additional European language (Spanish or German).
Success in Role
Sight data pipelines are reliable, monitored, documented and cost-effective, with data quality visibly managed.
Onboarding a new plant or asset group to the data platform becomes a configuration exercise, not a bespoke build.
Data scientists and engineers can access trustworthy, reproducible datasets without bespoke intervention.
A consistent common data model and asset hierarchy is in place across Sight products, supporting global replication.
Model performance is supported by demonstrable data lineage, labelling quality and drift monitoring.
Data governance, access control, privacy and cybersecurity requirements are met across OT and IT boundaries.
Data engineering standards and patterns you have defined are visibly used by other engineers across both product families.
Engineers you mentor demonstrably grow in autonomy and quality on data work; you are not a permanent bottleneck.
No critical dataset or pipeline depends on a single person.
Ford of Mexico is committed to being an equal opportunity employer. We believe that a diverse and inclusive workplace is essential to our success. We do not discriminate based on race, religion, color, national origin, gender, sexual orientation, age, marital status, pregnancy, veteran status, or disability status.
Our focus is to build the best team. If you think you can bring value to Ford, love to collaborate, prioritize and aim to deliver excellence in everything you do, we encourage you to apply!
We thank all candidates for their interest, but only those selected for an interview will be contacted.
#LI-NG2
#FordMexico
