Senior MLOps Engineer (Python)

Candidate's Portrait

Senior hands-on MLOps Engineer with strong expertise across MLOps, DevOps/SRE, software engineering, and production machine learning systems. The candidate should have proven ownership of ML platforms and be able to design, build, operate, and continuously improve the full lifecycle of ML and Generative AI solutions in production.
Strong engineering background is required, including Kubernetes, CI/CD, Infrastructure as Code, Python, observability, scalable production services, and ML-specific infrastructure. The candidate should be comfortable making architectural decisions, defining engineering standards, collaborating with Platform Engineering, and taking responsibility for reliability, performance, security, governance, and cost efficiency of AI workloads.
Hands-on LLMOps and Generative AI experience is highly valuable, especially production experience with RAG, vector databases, model gateways, prompt management, evaluation, guardrails, agentic systems, and LLM cost optimization. Strong communication skills are required to explain technical trade-offs to engineers, architects, and business stakeholders.

Must-haves

5+ years of strong engineering experience across Software Engineering, DevOps/SRE, and MLOps.
Proven ownership and operation of Machine Learning systems in production.
Strong hands-on Python development at production engineering level, including typed, tested, packaged, and reviewed code.
Deep hands-on experience with Docker and Kubernetes.
Practical Kubernetes experience with resource management and GPU-based ML workloads.
Strong experience designing and operating end-to-end MLOps pipelines, including training, model validation, model registry, deployment, serving, monitoring, and retraining.
Strong CI/CD experience with GitLab CI, GitHub Actions, ArgoCD, or equivalent platforms.
Experience implementing automated quality gates within CI/CD pipelines.
Strong hands-on experience with MLOps platforms and tools such as MLflow, Kubeflow, Feast, BentoML, KServe, SageMaker, Vertex AI, Databricks, or comparable solutions.
Strong Infrastructure as Code experience with Terraform, Ansible, or equivalent tools, including module and state management.
Experience building scalable production services with structured logging, API contracts, configuration management, and centralized secrets.
Strong experience with PostgreSQL, caching solutions, and asynchronous/message-driven processing.
Experience with production observability, including metrics, logging, error tracking, alerting, and incident response.
Experience with ML-aware monitoring, model performance monitoring, data quality, and model drift detection.
Experience with automated model testing, including data validation, regression testing, training-serving consistency, and evaluation gates.
Experience with model versioning, dataset versioning, experiment tracking, and reproducible ML workflows.
Experience implementing progressive delivery approaches such as canary releases, shadow testing, blue-green deployments, and automated rollback.
Experience designing automated retraining workflows based on schedules, events, or model/data drift.
Strong understanding of production performance, availability, latency, throughput, scalability, and cost optimization.
Proven experience collaborating with Platform Engineering, DevOps, Infrastructure, Security, and ML Engineering teams.
Ability to participate in architecture and roadmap discussions and translate ML infrastructure requirements into technical solutions.
Strong communication skills and ability to clearly explain technical trade-offs to technical and non-technical stakeholders.
Hands-on approach with strong architecture ownership and engineering accountability.

"Customer location: Saudi Arabia
❗️❗️❗️❗️Candidates locations: All except Russia
❗️❗️❗️❗️Language: English B2
Estimated start date: ••••••••
Duration: 3 months"

❗️❗️❗️Писать ••••••••⚡️⚡️⚡️⚡️ С РЕЗЮМЕ В КОТОРОМ УКАЗАН ВОЗРАСТ И ЛОКАЦИЯ НАХОЖДЕНИЯ. Укажите 🆔 вакансии❗️ Без этой информации отклики не рассматриваются ❗️❗️