Apply now
Master'sAutonomous, ADAS & Software-Defined Vehicle Applications open

Master’s in MLOps Engineering for AD/ADAS

Online master's programme of 1,900 hours in MLOps engineering for AD/ADAS, structured in 12 modules covering data infrastructure, curation and labelling, training at scale, model evaluation, in-vehicle deployment, fleet monitoring, ML CI/CD and compliance (ISO/PAS 8800, ISO 21448, ISO 26262), with an assigned technical tutor.

Duration1900 h
ModeOnline
LanguageEnglish
Start datesSet by the faculty
AwardSEIUM Institute diploma
Enrol now View syllabus Ask an advisor
Virtual lab · try it before you enrol

Take the controls.

Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.

SIM · LIVE ●
Learning outcomes

What you will be able to do

01

Design data infrastructures for test fleets.

02

Apply data curation and selection techniques.

03

Manage labelling workflows with quality control.

04

Train and evaluate models at scale in a reproducible way.

05

Deploy optimised models on automotive hardware.

06

Monitor models in the fleet and detect drift.

07

Develop an MLOps project for ADAS.

Profile

Who it's for

  • MLOps engineers moving into automated driving.
  • Driving data engineers managing fleet data pipelines.
  • Machine learning engineers who want to take perception models into production.
  • Data platform managers in automotive organisations.
  • Embedded deployment engineers optimising models for vehicle hardware.
  • Data quality managers responsible for datasets and labelling.
  • Engineering or science graduates seeking to enter the automotive sector.
Standards & tools

What you'll work with

  • ISO/PAS 8800 (safety and artificial intelligence in road vehicles)
  • ISO 21448 (SOTIF – safety of the intended functionality)
  • ISO 26262 (functional safety of road vehicles)
  • Regulation (EU) 2024/1689 (AI Act) (EU artificial intelligence regulation)
  • ISO/IEC 5338 (AI system life-cycle processes)
  • ISO/IEC 42001 (AI management systems)
  • UN ECE R156 (software updates and update management systems)
  • Regulation (EU) 2016/679 (GDPR) (EU data protection regulation)
  • PyTorch
  • MLflow
  • Kubeflow
  • DVC
  • NVIDIA TensorRT
Syllabus

12 modules · 1900 h

  1. 1.1Model life cycle in ADAS
  2. 1.2Differences between general and automotive MLOps
  3. 1.3The data engine and improvement loop
  4. 1.4Safety and traceability requirements
  5. 1.5Data volume in test fleets
  6. 1.6Roles in an MLOps team
  7. 1.7MLOps maturity by levels
  8. 1.8Industry reference cases
  9. 1.9AI infrastructure costs
  10. 1.10Case study: analysis of a manufacturer's data engine
  1. 2.1Data ingestion from test vehicles
  2. 2.2rosbag, MCAP and Parquet formats
  3. 2.3Data lake and lakehouse architectures
  4. 2.4Object storage at scale
  5. 2.5Data catalogues and metadata
  6. 2.6Indexing by scenarios and events
  7. 2.7Physical and network transfer
  8. 2.8Storage and retention costs
  9. 2.9Security and access control
  10. 2.10Lab: ingesting MCAP into a lakehouse
  1. 3.1Hard case mining
  2. 3.2Semantic search with embeddings
  3. 3.3Active learning for selection
  4. 3.4Novelty detection in fleet data
  5. 3.5Balancing classes and scenarios
  6. 3.6Sequence deduplication
  7. 3.7In-vehicle capture triggers
  8. 3.8ODD coverage in datasets
  9. 3.9Data value metrics
  10. 3.10Practical: embedding search with CLIP
  1. 4.12D and 3D annotation workflows
  2. 4.2Auto-labelling with offline models
  3. 4.3Point cloud labelling
  4. 4.4Labelling quality control
  5. 4.5Consistent annotation guidelines
  6. 4.6Managing labelling suppliers
  7. 4.7Inter-annotator agreement metrics
  8. 4.8Correction of label errors
  9. 4.9Costs and productivity
  10. 4.10Workshop: 3D auto-labelling with human review
  1. 5.1Distributed data and model training
  2. 5.2GPU clusters and scheduling
  3. 5.3Experiment tracking with MLflow
  4. 5.4Hyperparameter management
  5. 5.5Reproducibility of training runs
  6. 5.6Data versioning with DVC
  7. 5.7Multi-task perception training
  8. 5.8Optimisation of computing costs
  9. 5.9Checkpointing and fault tolerance
  10. 5.10Lab: distributed training with PyTorch
  1. 6.1Perception metrics by class and distance
  2. 6.2Evaluation by critical scenarios
  3. 6.3Frozen test sets
  4. 6.4Detection of regressions between versions
  5. 6.5Closed-loop evaluation
  6. 6.6Systematic error analysis
  7. 6.7Uncertainty calibration
  8. 6.8Evaluation of fairness and bias
  9. 6.9Evaluation reports
  10. 6.10Practical: comparison of two detector versions
  1. 7.1Exporting models to ONNX
  2. 7.2Compilation with TensorRT
  3. 7.3Quantisation and mixed precision
  4. 7.4Validation of numerical equivalence
  5. 7.5Packaging the model into software
  6. 7.6Version management on the ECU
  7. 7.7OTA model updates
  8. 7.8Testing on target hardware
  9. 7.9Shadow mode deployment
  10. 7.10Workshop: deployment with TensorRT and validation
  1. 8.1Model performance telemetry
  2. 8.2Data drift detection
  3. 8.3Concept drift detection
  4. 8.4Confidence monitoring
  5. 8.5Data collection triggers
  6. 8.6Model health dashboards
  7. 8.7Alerts and response
  8. 8.8Feedback into training
  9. 8.9Fleet data privacy
  10. 8.10Lab: drift detection with Evidently
  1. 9.1Automated training pipelines
  2. 9.2Orchestration with Kubeflow and Airflow
  3. 9.3Continuous integration of models
  4. 9.4Automated model testing
  5. 9.5Quality gates for release
  6. 9.6Model registry
  7. 9.7Infrastructure as code
  8. 9.8Environment management
  9. 9.9Delivery metrics
  10. 9.10Practical: ML pipeline with Kubeflow
  1. 10.1ISO/PAS 8800 requirements for the life cycle
  2. 10.2Traceability between data, model and release
  3. 10.3Evidence for SOTIF
  4. 10.4Change management for models
  5. 10.5AI Act requirements
  6. 10.6Technical documentation of models
  7. 10.7Cybersecurity of the ML pipeline
  8. 10.8Audits of MLOps processes
  9. 10.9Tool qualification
  10. 10.10Case study: traceability of a perception release
  1. 11.1Structure of data and ML teams
  2. 11.2In-house platform versus suppliers
  3. 11.3Cloud costs versus own infrastructure
  4. 11.4FinOps applied to ML
  5. 11.5Managing data suppliers
  6. 11.6Productivity indicators
  7. 11.7Collaboration with safety teams
  8. 11.8Scaling the organisation
  9. 11.9Platform roadmap
  10. 11.10Workshop: cost estimation for a data engine
  1. 12.1Definition of the model and the function
  2. 12.2Design of the data infrastructure
  3. 12.3Curation and labelling strategy
  4. 12.4Training and evaluation pipeline
  5. 12.5In-vehicle deployment strategy
  6. 12.6Fleet monitoring
  7. 12.7Traceability and compliance
  8. 12.8Conclusions and roadmap
  9. 12.9Thesis report
  10. 12.10Master's Thesis: defence before a panel
Methodology, software & labs

How you'll learn

100% online training on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. There are no fixed timetables: you progress at your own pace within each intake, making it compatible with work.

Each module combines theory, guided reading of the original standard and applied exercises. The working software is PyTorch, MLflow, Kubeflow, DVC and NVIDIA TensorRT, in free, academic or evaluation versions. Testing is studied through real procedures: hard case mining in fleet data, evaluation of regressions between model versions and shadow mode deployment. Assessment is continuous for each module, plus the final Master's Thesis.

Standards covered: ISO/PAS 8800, ISO 21448, ISO 26262, Regulation (EU) 2024/1689 (AI Act), ISO/IEC 5338, ISO/IEC 42001, UN ECE R156, Regulation (EU) 2016/679 (GDPR).

Assigned technical tutor:every module ends with a deliverable that your tutor marks individually.
Standards-based learning:guided reading of ISO/PAS 8800, ISO 21448 and ISO 26262 applied to real cases, not as a theoretical annex.
Industry software:PyTorch, MLflow, Kubeflow, DVC and NVIDIA TensorRT, in free, academic or evaluation versions.
Master's Thesis with defence:a complete MLOps project for an ADAS function, from data infrastructure and labelling to deployment, fleet monitoring and traceability, defended before a panel.
Flexible format:100% online, no fixed timetables, compatible with full-time work.
Capstone project

Build it end to end

Your final project follows the real engineering workflow, step by step, reviewed by your tutor.

Fleet data ingestion
Ingest MCAP test-vehicle data into a lakehouse with scenario indexing.
Curation and auto-labelling
Select hard cases by embedding search and auto-label them in 3D with human review.
Training and evaluation
Run distributed training with PyTorch and compare two detector versions for regressions.
Deployment and monitoring
Deploy the model with TensorRT and detect drift in fleet data with Evidently.
Master's Thesis
A complete MLOps project for an ADAS function, from data infrastructure and labelling to deployment, fleet monitoring and traceability, defended before a panel.
Career opportunities

Where this takes you

MLOps engineerDriving data engineerMachine learning engineerData platform managerEmbedded deployment engineerData quality manager
Admissions & scholarships

Requirements

  • A university degree in engineering, science or architecture, or equivalent demonstrable professional experience.
  • An internet connection and a computer capable of running the programme software.
  • English sufficient to read technical standards.

Ready to start?

Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.

Apply now Scholarships
FAQ

Frequently asked questions

The programme totals 1,900 hours over 12–14 months. You set your own pace within each intake, and your tutor helps you plan the workload around your job.

You work with PyTorch, MLflow, Kubeflow, DVC and NVIDIA TensorRT, using free, academic or evaluation versions. Your tutor indicates the free alternative wherever one exists.

A SEIUM Institute diploma, awarded on passing the continuous assessment and the Master's Thesis. It is SEIUM's own qualification, not an official state degree; its recognition depends on the programme completed and the regulations of each country.

Yes. Modules passed in a course or postgraduate diploma in the same area are recognised in the higher programme. Ask about the recognition policy when you enrol.

A degree in engineering, science or architecture (or equivalent professional experience), a computer able to run the software and English sufficient to read technical standards.

Start dates are set by SEIUM's faculty for each intake. Our admissions team will confirm the next available date when you apply.

Keep exploring

Related programmes

See all in Autonomous, ADAS & Software-Defined Vehicle
Master's6,610 $

Master’s in Semantic Mapping and Risk-Prioritised Maps Engineering

Online master's programme of 1,900 hours in semantic cartography and risk-prioritised maps engineering, structured in 12 modules covering semantic map models and ontologies, semantic segmentation, road…

1900 hOnlineAutonomous, ADAS & Software-Defined Vehicle
Master's7,730 $

Master’s in Heterogeneous Computing for AD (CPU/GPU/NPU)

Online master's programme of 1,900 hours in heterogeneous computing for autonomous driving (CPU/GPU/NPU). 12 modules covering GPU programming, neural network inference, task scheduling, memory, power and…

12–14 monthsOnlineAutonomous, ADAS & Software-Defined Vehicle
Master's6,610 $

Master’s in Teleoperation and Control Centre Engineering

Online master's programme of 1,900 hours in teleoperation and control centre engineering, structured in 12 modules covering communications for teleoperation, remote driving and assistance, control centre…

1900 hOnlineAutonomous, ADAS & Software-Defined Vehicle
Master’s in MLOps Engineering for AD/ADASEnrol
Related programmes

More in Autonomous, ADAS & Software-Defined Vehicle

Master’s in Robust Perception in Adverse Conditions EngineeringView programme →Master’s in Traffic and Proving Ground Digital Twins EngineeringView programme →Master’s in Heterogeneous Computing for AD (CPU/GPU/NPU)View programme →Master’s in OTA, Diagnostics & Operations EngineeringView programme →Master’s in Software-Defined Vehicle DevOps EngineeringView programme →Master’s in Off-Highway Autonomy EngineeringView programme →

All Autonomous, ADAS & Software-Defined Vehicle programmes → · How to apply · Fees & financing · Scholarships

Scroll to Top
SEIUM – Institute of Advanced Engineering
Powered by  GDPR Cookie Compliance
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.