Machine Learning Ops Engineer
PAR Technology3 hours ago
Remote, United StatesMid Level / Senior
H1B sponsor
Base Salary
$125k - $160k/yr
Responsibilities
- Build and maintain automated, tested ML and data pipelines in partnership with data science teams.
- Optimize Spark jobs and develop reliable distributed data-processing workflows.
- Implement CI/CD workflows, automated unit and integration tests, and branch-based promotion across environments.
- Orchestrate workflows with retries, backfills, SLA monitoring, and alerting.
- Manage experiment tracking, model registries, lineage, and stage transitions.
- Develop monitoring and observability for data quality, schema drift, and input/output distribution drift.
- Build infrastructure as code, secrets management, reproducible Python environments, and cloud-native production systems.
- Serve processed outputs to downstream systems such as Redshift for BI, activation, and attribution.
- Support exploratory data analysis, data-quality investigations, feature pipelines, training-serving consistency, and champion/challenger evaluation.
- Build lightweight interfaces and structured data feeds for LLM-based agent workflows while maintaining appropriate data contracts.
- Use AI tools to accelerate coding, documentation, log analysis, dataset exploration, and evaluation-case generation.
- Work carefully with PII-bearing datasets and support privacy-conscious entity-resolution systems.
Requirements
- 2–4 years of experience building production data or ML systems.
- Bachelor’s degree in Computer Science, Data Engineering, or a related field, or equivalent practical experience.
- Experience building and maintaining automated, tested ML or data pipelines.
- Experience with Git-based workflows, automated unit and integration testing, and CI/CD promotion across development, staging, and production.
- Experience with workflow orchestration using Airflow, Step Functions, or Databricks Workflows.
- Familiarity with MLflow or equivalent experiment tracking and model registries.
- Exposure to data-quality and observability tools such as Great Expectations, Deequ, or Datadog.
- Experience with infrastructure as code, secrets management, and reproducible Python environments.
- Production experience with AWS and working knowledge of Spark performance, partitioning, joins, skew, shuffle, and spill.
- Experience serving processed outputs to downstream systems such as Redshift.
- Experience building feature pipelines and maintaining consistency between training and serving.
- Fluency across the Python data science stack, including pandas, scikit-learn, and PySpark.
- Applied statistics knowledge covering probability, hypothesis testing, sampling, and evaluation metrics.
- Ability to perform exploratory data analysis, investigate data quality, read unfamiliar ML code, and collaborate with data science teams.
- Interest in LLM-based agent data integration, tool definitions, retrieval patterns, and structured data interfaces.
- Experience with agentic frameworks such as LangChain or LlamaIndex is preferred.
- Exposure to PII-based record linkage, probabilistic matching, graph or clustering-based entity resolution, blocking strategies, and cloud cost optimization is preferred.
- Strong communication, collaboration, ownership, adaptability, and accountability.
Benefits
- Remote work is available in New York, Texas, Philadelphia, Florida, North Carolina, Minnesota, Colorado, Georgia, and Illinois.
- The role includes collaboration with cross-functional teams and exposure to cloud-native SaaS, ML operations, data engineering, and AI-agent integration work.
Tech Stack
Amazon RedshiftApache AirflowApache SparkAWSDatadogDockerGraphQLMLflowPandaspytestPythonscikit-learnSnowflakeTerraform
Categories
Data EngineeringML Engineering
About PAR Technology
PAR Technology (NYSE: PAR) builds software and hardware for multi‑unit restaurants and retailers, including point of sale, digital ordering, loyalty, payments, and back‑office systems. The company sells a SaaS platform with integrated devices and services used by over 100,000 restaurant locations worldwide. Founded in 1968 and headquartered in New Hartford, NY, PAR focuses on enterprise operators and franchises that need to manage performance across many locations.