MLOps Engineer

About the Role

We are seeking a mid- to senior-level MLOps Engineer to contribute to building the intelligent foundation of our AI platform. This role surpasses typical MLOps duties that focus solely on pipeline maintenance and deployment, involving the creation of core infrastructure to empower sophisticated AI workflows accessible via a user-friendly, self-service platform. Features include model deployment, training, notebooks, functions, UI-driven agent setup, and RAG pipelines.

This position lies at the crossroads of platform engineering, machine learning infrastructure, distributed systems, and improving developer experience.

Key Responsibilities

• Architect and implement the infrastructure supporting the Intelligence layer.
• Establish dependable, automated workflows encompassing model training, deployment, lifecycle administration, and inference.
• Create scalable foundations to allow users to build, configure, and manage AI agents and retrieval-augmented generation (RAG) pipelines via the platform’s UI.
• Develop platform functionalities linked to managed notebooks, functions, experiments, training tasks, model registries, and serving endpoints.
• Enhance model serving capabilities such as observability, version control, evaluation, promotion, and rollback.
• Optimize GPU-based inference and training deployments focusing on performance, dependability, and cost-effectiveness.
• Research effective deployment techniques for models across centralized GPU infrastructure and edge devices.
• Automate workflows traditionally requiring manual MLOps work, emphasizing secure, user-friendly self-service features.
• Collaborate with backend, product, and AI teams to transform complex infrastructure into intuitive platform features.
• Contribute to defining standards for security, multi-tenancy, resource isolation, model governance, and operational reliability.

Technology Stack

• Kubernetes and Knative
• KServe and vLLM
• MLflow
• LangFuse
• GPU infrastructure and model serving workloads
• RAG architectures, vector retrieval, agent workflows, and large language model (LLM) applications

Our backend intelligence is implemented in Rust, making Rust proficiency or interest highly advantageous.

Candidate Profile

• Demonstrated expertise in MLOps, AI platform engineering, machine learning infrastructure, or distributed systems.
• Practical experience with Kubernetes, including deploying and managing stateful, training, notebook, serverless, or GPU-intensive workloads.
• Experience designing and operating ML platforms supporting training, experimentation, notebooks, feature/data workflows, model registries, and production model serving.
• Proficiency with model serving frameworks such as KServe, vLLM, Triton, or Ray Serve.
• Familiarity with ML lifecycle tools like MLflow, experiment tracking, model registries, and CI/CD or GitOps practices tailored to ML systems.
• Skill in optimizing training and inference for latency, throughput, availability, and cost efficiency.
• Understanding of GPU scheduling, resource allocation, model quantization, batching strategies, autoscaling, and managing multi-model serving.
• Experience with RAG systems, LLM applications, AI agents, or their supporting infrastructure.
• Solid software engineering background combined with a product-focused mindset for platform development.
• Capability to simplify complex operational workflows to enhance user reliability and accessibility.

Preferred Qualifications

• Experience or interest in Rust for backend systems development.
• Familiarity with notebook platforms like JupyterHub, Kubeflow Notebooks, or VS Code Server.
• Background in creating secure, scalable execution environments for user-defined functions or jobs.
• Expertise in deploying AI workloads on edge devices.
• Knowledge of model optimization methods such as quantization, compilation, pruning, and hardware-aware serving.
• Experience working with multi-tenant AI platforms, including security and model governance measures.
• Understanding observability and evaluation tools for ML and LLM systems.

Success Criteria

Your efforts will transition our AI infrastructure from manual operation to a sophisticated platform enabling users to train, assess, deploy, and run models, operate managed notebooks, execute functions, build agents, and configure RAG workflows—while the platform manages most operational complexities.

Data Privacy Consent

By applying, you consent to the collection, processing, and retention of your personal information solely for recruitment and evaluation purposes by the company.