Senior ML Ops Engineer

Confido•New York, NY

9d•$210,000 - $300,000•Onsite

About The Position

Confido is hiring its first dedicated owner of its ML platform. The AI/ML team already deploys document-understanding, forecasting, and agentic systems in production, and this role will own the infrastructure layer that makes models and agents reliable and cost-efficient at scale, including workloads involving hundreds of thousands of documents and heavy LLM/VLM use. The position is based in New York, NY, with relocation support. The role owns ML pipelines from experimentation through production, along with infrastructure for training, inference, and agentic workloads. It will establish reproducible environments and streamlined paths from prototypes to production, build cloud infrastructure as code and CI/CD, optimize inference and forecasting performance and cost, and maintain data interfaces with data engineering. It will also make reliability, observability, security, and privacy foundational, while ensuring model and agent quality is measured in production through online evaluations and human review.

Requirements

At least 5 years of experience in MLOps, ML platform, AI infrastructure, or platform engineering involving production ML systems.
Ability to work across software and infrastructure, including writing production code and building cloud infrastructure.
Experience driving a production pipeline end to end and explaining its architecture, security and cost trade-offs, and potential improvements.
Deep understanding of cloud infrastructure, distributed data systems, and Infrastructure as Code; ability to build an environment from scratch, configure CI/CD, and operate containerized workloads in production independently.
Strong Python skills and comfort working in a production application codebase such as Ruby or Java.
Treat monitoring, security, and cost as core considerations.
High ownership in a fast-moving startup environment and experience productionizing work built by research or AI teams.

Nice To Haves

Experience with LLMOps tooling such as tracing, prompt/version management, or evaluation harnesses.
Inference optimization experience with vLLM, ONNX, or TensorRT, and knowledge of GPU or spot-instance economics.
Experience with ML platform and orchestration tools such as MLflow, BentoML, Ray, or Airflow.
Experience with large-scale data systems such as Snowflake or Kafka and vector databases.
Experience with managed ML services such as Bedrock, SageMaker, or Vertex AI.
Experience with multimodal or generative AI in production.

Responsibilities

Own ML pipelines end to end, from experimentation to production, and the infrastructure for training, inference, and agentic workloads.
Create reproducible environments and efficient paths from prototype to production for the AI/ML team.
Build the cloud foundation using Infrastructure as Code and implement CI/CD for safe ML deployment.
Serve and optimize inference and forecasting workloads for latency, throughput, and cost, and optimize their data streams.
Own the data interface with data engineering, providing appropriate data to models and agents and writing outputs back into platform data systems.
Make reliability, observability, security, and privacy standard across the platform.
Measure model and agent quality in production through online evaluations and human-in-the-loop review.

Benefits

Equity in the company.
Fully paid employee health coverage through Aetna.
Dental and vision coverage through Guardian.
Carrot Fertility Pro fertility and family-forming support.
12 weeks of paid parental leave.
Unlimited paid time off and regular four-day holiday weekends.
401(k) through Vestwell.
Paid relocation support to New York City.
Equipped workspace, including a laptop, monitor, keyboard, and a $200 setup-personalization stipend.
Catered Friday lunches, team dinners, and unlimited coffee and snacks featuring products from the brands Confido works with.