Real-Time ML Engineer on Databricks & Spark

EPAM Systems; Turkey; On-site; TRY 3,949,000 - 6,417,000; Full time

Benefits offered by this job

Private health insurance

Extra leave days

Referral bonuses

IT Equipment

English courses

Job summary

EPAM Systems seeks a hands-on AI/ML Engineer to design and deploy optimized feature engineering and real-time inference pipelines, including Spark Streaming and batch training. You will work with Databricks, PySpark, Python, and SQL to process structured and unstructured data at scale and ensure low-latency deployment in production.

Applicants should have 3+ years of experience, strong knowledge of Spark Streaming, and English at B2 level or higher, with a track record of production-grade ML

Qualifications

• 3+ years of experience building, deploying, and operating offline and online feature engineering pipelines.
• Strong experience with model training and real-time inferencing pipelines.
• Hands-on expertise in Databricks, Spark/PySpark, Python, and SQL.
• Experience processing large-scale unstructured data.
• Strong knowledge of Spark Streaming.
• Experience with real-time feature generation and low-latency model inference.
• Experience building and supporting batch pipelines for model training and retraining.
• Proven track record of running production-grade ML systems.
• English proficiency at B2 level or higher.

Responsibilities

• Design and optimize feature engineering pipelines for both batch and real-time workloads.
• Build, deploy, and maintain scalable model training and inference pipelines.
• Process structured and unstructured data at scale.
• Develop streaming solutions using Spark Streaming.
• Enable real-time feature generation and model serving.
• Ensure reliability, scalability, and performance of ML solutions in production.

Skills

Databricks

Spark/PySpark

Python

SQL

Spark Streaming

Tools

MLflow

AWS

Job description

EPAM Systems seeks a hands-on AI/ML Engineer to design and deploy optimized feature engineering and real-time inference pipelines, including Spark Streaming and batch training. You will work with Databricks, PySpark, Python, and SQL to process structured and unstructured data at scale and ensure low-latency deployment in production.

Applicants should have 3+ years of experience, strong knowledge of Spark Streaming, and English at B2 level or higher, with a track record of production-grade ML