Principal MLOps Engineer

The Principal MLOps Engineer in the Acxiom Data Science and Machine Learning team will spearhead the development of an MLOps platform to support the development and lifecycle of Acxiom’s modeled propensities. This role integrates software engineering, AI/ML engineering, data proficiency, and MLOps experience to build a state-of-the-art MLOps solution that can power our model product builds, and other complex marketing activities/
As a Principal MLOps Engineer, you will collaborate with the MLOps engineering lead to modernize and operationalize Acxiom's Machine Learning platform and its machine learning pipelines, which process terabytes of data. Your responsibilities will include defining requirements, partnering with the Architecture Center of Excellence to establish the new MLOps platform architecture, and leading the hands-on development of MLOps pipelines capable of supporting a large portfolio of ML models and their lifecycles.
This role can be located almost anywhere in the U.S.

What You Will Do:

• Partner with the MLOps Engineering leader, Architecture and data science teams to design and develop hyperscale ML engineering and MLOps solutions and pipelines.
• Partner with resources across US, Europe and Asia to own development and modernization activities
• Assess current state of MLOps and  AI/ML/GenAI capabilities, identify gaps, and design target-state architectures to support ongoing modeled product builds, innovation, revenue growth, and operational excellence.
• Own the development of new modernized MLOps infrastructure and migration of existing data products to new infrastructure
• Develop automated AI and ML workflows and end-to-end pipelines for data preparation, training, deployment, and monitoring, ensuring the quality of architecture and design of our ML systems and data infrastructure.
• Collaborate with Data Scientists, Product Owners, ML Engineers, and Software Engineers to design and deliver ML solutions, promote models and associated MLOps pipelines into production.
• Leverage AI to develop GenAI-powered solutions to complement our data science and product build capabilities.
• Lead transformational initiatives to bridge the gap between current and desired AI/ML capabilities, collaborating with cross-functional teams to ensure successful implementation.
• Establish governance frameworks and decision criteria for AI/ML and GenAI projects, ensuring adherence to industry standards, regulatory requirements, Responsible AI principles, and Acxiom/IPG’s architectural guidelines.
• Partner with Architecture COE to create and maintain reference architectures, patterns, and best practices for the AI/ML lifecycle and its integration within Acxiom’s enterprise ecosystem.
• Own the ongoing support of this modernized platform once its built and operationalized developing new features and capabilities.
• Lead the ongoing technology evaluation and process improvements to drive experimentation, model development, and MLOps at scale.
• Lead and drive standardization of LLM onboarding processes, RAG pipelines, and application development.
• Conduct periodic architecture reviews and risk assessments for proposed AI/ML solutions, ensuring they meet security, scalability, and interoperability requirements.
• Maintain high reliability of machine learning pipelines in production environments, ensuring minimal downtime and optimal performance.

What You Will Have:

• 10+ years of experience in enterprise architecture, with a focus on AI/ML integration and transformation projects.
• 8+ years of professional experience in software development.
• Bachelor’s Degree in Computer Science or Associate Degree & 8+ years of development experience or equivalent experience.
• Strong computer science fundamentals in object-oriented design, data structures, algorithm design, problem-solving, and complexity analysis.
• Proficiency in at least two modern programming languages such as Java, C++, C, or Python.

Preferred Skills:

• 10+ years of experience in MLOps and ML Platform engineering, especially architecting scalable MLOps infrastructure and big data systems.
• Proven experience building ML platforms that can run large-scale model training & inferences (Trillions of inferences).
• Proven experience with ML libraries like H2O, SparkML, scikit-learn, and deep learning frameworks (PyTorch, TensorFlow, etc.).
• 8+ years of experience deploying ML solutions in Java, C/C++.
• Databricks ML Professional Certification or equivalent is required.
• 8+ years of experience optimizing Spark workloads with deep Spark troubleshooting experience.
• 6+ years of architecting solutions using Databricks, with strong experience using Mosaic AI, Unity Catalog, MLflow, workflow orchestration, and other Databricks native MLOps capabilities.
• At least 2+ years of experience in GenAI, including technical familiarity with at least two of the following: OpenAI API, Bedrock API, Vertex API, LangGraph, or other agentic frameworks.
• Exceptional attention to detail and proven ability to manage multiple competing priorities simultaneously.
• Experience with MLOps and orchestration tools such as Airflow, Kubeflow, DAGster, Optuna, or MLflow.
• Strong CI/CD experience using tools like Terraform, Jenkins, and CloudFormation templates.
• Experience with operationalizing and migrating ML models into production at scale.
• Experience developing large-scale model inference solutions using parallel execution frameworks with Spark, EMR, or Databricks.
• Experience developing complex orchestration and MLOps pipelines stitching together large volumes of data for training and scoring.
• Experience with Large Language Models, fine-tuning, and deployment frameworks using Hugging Face capabilities or cloud provider solutions such as Amazon Bedrock or Vertex AI Model Garden.
• Familiarity with vector databases such as Pinecone or ChromaDB.
• Experience in CI/CD/DevOps, Deployment and Automation Tools – CI/CD, Jenkins, Terraform, Cloud Formation Template or similar.
• Proficiency with Apache Spark, EMR/DataProc, and cloud-based tools (Snowflake, Redshift, EMR, Glue, Step Functions, Lambda, Step functions, AWS Batch, or similar).
• Excellence in technical communication with scientists and engineers.
• At least 6+ years of Database (SQL) experience and Linux experience.
• At least 10+ years of AWS infrastructure experience - Cloud Run, App Server, RDS, S3, EC2, EMR or equivalent GCP experience.

What will set you apart:

• Databricks Certification, Snowflake Certification or equivalent.
• LangGraph, Databricks MLflow experience, Docker experience, Kubernetes experience

#GD17

Primary Location City/State:

Homebased - Conway, Arkansas

Additional Locations (if applicable):

Acxiom is an equal opportunity employer, including disability and protected veteran status (EOE/Vet/Disabled) and does not discriminate in recruiting, hiring, training, promotion or other employment of associates or the awarding of subcontracts because of a person's race, color, sex, age, religion, national origin, protected veteran, military status, physical or mental disability, sexual orientation, gender identity or expression, genetics or other protected status.

Attention California Applicants:  Please see our CCPA/CPRA Privacy Act notice here.

Attention Colorado, California, Connecticut, Maryland, Nevada, New Jersey, New York City, Ohio, Rhode Island, and Washington Applicants: This position is not located in the aforementioned locations but applications for remote work may be considered. For information about this role under state or local equal pay or pay transparency laws, please contact [email protected].