Senior ML Engineer

Undelucram.ro on behalf of:

SoftServeAbout The Role

In this role you will engineer and productionize end-to-end ML systems — from data pipelines and LLMOps infrastructure to agentic multi-agent workflows — as part of SoftServe's AI and Data Science Center of Excellence, a community of over 170 AI/ML experts. You'll work at the intersection of applied research and real-world delivery, collaborating with data scientists, engineers, and clients to bring cutting-edge NLP, RAG, and multimodal AI solutions to production scale.

Responsibilities

• Design and implement end-to-end ML pipelines — from data ingestion and feature engineering to model training, optimization, and production deployment

• Build and maintain LLMOps pipelines using MLflow, Langfuse, LangSmith, or Weights & Biases to enable model observability, reproducibility, and prompt versioning

• Collaborate with Data Scientists, Engineers, and clients to translate business requirements into production-ready ML solutions for NLP, RAG systems, and multimodal models

• Develop and orchestrate agentic systems and multi-agent workflows using frameworks such as LangGraph or CrewAI, supporting autonomous AI applications at scale

• Enhance and manage ML infrastructure including CI/CD/CT pipelines, cloud environments on AWS, Azure, or GCP, data stores, monitoring, and security

• Integrate and package ML services into real applications, ensuring they meet reliability and maintainability standards for production use

• Operate workflow orchestration tools such as Databricks Jobs/Workflows, Kubeflow, or Airflow to automate and monitor ML pipeline execution

Requirements

• Minimum 3 years of hands-on experience building and deploying real-world ML solutions in production

• Master's degree in Computer Science or a related field

• Strong Python proficiency across the core data science and ML ecosystem, including model development, packaging, and service integration

• Advanced experience with LLMOps, AgentOps, and experiment tracking tools such as MLflow, Langfuse, LangSmith, and Weights & Biases

• Solid knowledge of CI/CD/CT practices for ML systems and workflow orchestration tools such as Databricks Workflows, Kubeflow, or Airflow

• Proven experience with cloud-based AI/ML services on AWS, Azure, or GCP

• Working knowledge of agentic AI frameworks, including LangGraph, CrewAI, or similar tools for building autonomous and multi-agent systems

• Upper-intermediate or higher proficiency in English, both spoken and written