> Markdown version of [/jobs/ext/4749619-saas-e-commerce-client-3-senior-mlops-engineer-data-ml-platform](https://www.wearedevelopers.com/jobs/ext/4749619-saas-e-commerce-client-3-senior-mlops-engineer-data-ml-platform). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # SaaS e-commerce Client 3 | Senior MLOps Engineer Data & ML Platform - **Company:** SDS, INC. - **Location:** Atlanta, GA, United States - **Experience:** Expert - **Contract:** Permanent contract - **Skills:** Java (Programming Language), Application Programming Interfaces (APIs), Artificial Intelligence, Airflow, Amazon Web Services, Amazon S3, Big Data, Software as a Service, Continuous Integration, Data Validation, Information Engineering, Information Leak Prevention, Github, Identity and Access Management, Python (Programming Language), Operational Databases, Raw Data, Standard Sql, Azure Machine Learning, SQL Databases, Feature Store, Large Language Models, Snowflake, Grafana, Apache Spark, Agentic-AI, Apache Kafka, Spark Streaming, Data Management, Machine Learning Operations, Terraform, Data Pipelines, Docker, Databricks, Web Api - **Published:** November 16, 2026 - **Apply:** https://www.careerbuilder.com/job-details/saas-e-commerce-client-3-senior-mlops-engineer-data-ml-platform-atlanta-ga--e5efc5cf-a7cc-4141-9423-4f9e6e6c7119 ## About the Role eCommerce, Software as a Service (SaaS), Software as a Service (SaaS), eCommerce, Data Management, Application Programming Interface (API), Web Feeds, Data Quality, Root Cause Analysis, Data Modeling, * 5+ years in data engineering, MLOps or ML platform roles, with pipelines you've run in production * Production Spark: complex transformations, window functions, large joins, partitioning, and tuning for shuffles and skew. Databricks (jobs, Delta tables) is a big plus. * Production Snowflake: SQL, loading and unloading, and cost management * Large data volumes: hundreds of millions of rows a month, with incremental loads and backfills that keep run time and cost under control * Strong Python: clean, tested code; Spark UDFs including pandas UDFs; pandas and PyArrow * Data quality and monitoring: quality checks, drift and anomaly detection, and alerts people can act on * Deployment: GitHub Actions, Docker, AWS (S3, IAM) and proper secrets handling * Backend engineering: comfortable working in a service codebase, not only in notebooks and SQL * AI-first habits: you already use coding agents daily and look for what to automate * Fluent English and based in Georgia Nice to have: * MLflow and feature stores * Airflow or Databricks Workflows * Kafka or Spark Structured Streaming * Terraform * DBT * Scala for Spark * Reading Java * Time-based splits and preventing data leakage in training sets * Ad-tech or e-commerce data experience ## Description What you'll do: * Own the ML data pipelines: build and run batch pipelines on Spark and Databricks, from raw data to feature and training tables, including enrichment and backfills for late-arriving data. * Onboard new data sources: bring in internal databases, external APIs, partner feeds and web data, and match them to Falcon's own entities. * Use LLMs inside pipelines: add LLM steps to batch jobs, such as turning text into embeddings and structured attributes. * Keep training and serving consistent: make sure features match between training and the live inference service, which includes reading its Java code. * Monitor data and model inputs: run scheduled data quality checks and Grafana alerts that catch bad data before it reaches the model, and find the root cause when something breaks. * Ship like an MLOps engineer: deploy pipelines through CI/CD (GitHub Actions, Docker, AWS) with proper secrets handling, and keep run time and cost under control. * Work with AI agents: coding agents write most of Falcon's code. You'll decide where agents fit, then check their output on row counts, distributions and edge cases before anyone relies on it.