The posting
- High Technical Impact: A unique professional challenge to build and operate production ML systems and agentic workflows for a stable, market-leading global enterprise.
- Competitive Compensation: Competitive base salary and annual target bonus.
- Benefits Package: Cafeteria allowance and private health insurance package.
- Flexibility: Hybrid working model requiring 3 days of office presence (Budapest) and offering 2 days of Home Office per week.
Required Experience
- Industry Experience: 3+ years in machine learning engineering, applied ML, or software engineering with a meaningful ML focus and recent production delivery.
- Production Deployment: Production experience deploying and maintaining ML models in a serving environment handling real inference traffic beyond the notebook stage.
- Databricks & AWS Ecosystem: Working proficiency with the Databricks ML ecosystem (MLflow, Model Serving, Feature Store) and familiarity with supporting AWS services (SageMaker, S3, Lambda, Step Functions, ECS/EKS).
- LLMs & Agentic AI: Hands-on experience building applications with LLMs: RAG implementations, vector databases, prompt engineering, and API-based model integration (OpenAI, Anthropic, Bedrock, or equivalent). Production exposure to agentic AI patterns: multi-step orchestration, tool use, or workflow automation using LangChain, LangGraph, or similar frameworks.
- Software Engineering & Core ML: Strong Python proficiency and solid software engineering fundamentals: version control, testing, CI/CD, code review, and debugging production systems. Experience with at least one core ML framework (PyTorch, TensorFlow, scikit-learn, XGBoost) applied to production prediction tasks.
- Deployment & Infrastructure: Familiarity with containerized deployment (Docker, ECS, or Kubernetes) and infrastructure-as-code concepts.
Preferred Qualifications
- Experience with real-time inference serving and performance tuning: latency profiling, model optimization, caching, and scaling.
- Hands-on experience with evaluation frameworks for agentic systems: task success measurement, hallucination detection, and cost tracking.
- Familiarity with streaming data systems (Kafka, Kinesis, Spark Structured Streaming) as they relate to real-time feature computation and online inference.
- Experience with fine-tuning LLMs or training custom models on domain-specific data.
- Exposure to data governance practices: lineage tracking, access controls, and model documentation within Unity Catalog or equivalent.
- BS or MS in Computer Science, Machine Learning, or a quantitative field (practical production experience weighted equally).
- ML Pipeline Engineering: Build and maintain production ML pipelines — feature engineering, model training, evaluation, deployment, and monitoring on AWS and Databricks.
- Agentic AI Workflows: Implement and operate agentic AI workflows using frameworks such as LangChain, LangGraph, or equivalent, following architectural patterns established by senior engineers while contributing improvements based on production observations.
- Feature Engineering Pipelines: Develop and maintain feature engineering pipelines, ensuring data quality, freshness, and consistency across batch and real-time serving paths.
- MLOps Infrastructure: Operate and extend MLOps infrastructure: model CI/CD, experiment tracking, automated retraining, model versioning, and performance monitoring using MLflow and Databricks tooling.
- LLM Applications: Build and maintain RAG pipelines, vector database integrations, prompt management systems, and tool-use components for LLM-powered applications.
- Monitoring & Alerting: Implement model monitoring dashboards and alerting: tracking drift, latency, error rates, and cost, escalating anomalies with context.
- Code Quality & Production Operations: Write production-quality Python code with thorough testing, documentation, and adherence to team engineering standards. Participate in on-call rotations for ML systems, triaging production issues and implementing fixes with appropriate urgency.
- Cross-Team Collaboration: Collaborate with data engineers on upstream pipeline dependencies and with data scientists on translating research outputs into deployable services.
