About the Job

You'll be part of the AI Platform Team, which is currently taking shape. It's emerging within our DXP team (Developer Experience Platform), which has already built and operates a Kubernetes-based Internal Developer Platform. The new team is deliberately kept small, senior-led, and close to the existing platform expertise.

You'll build the path that takes AI into production here:

From prototype to productive application. You build the bridge from "sketched out with Claude in Python" to "running—observed and secured—in operations."
Selecting and integrating components. LLM gateway, container runtime, workflow and agent orchestration, observability, cost tracking. As a team, not alone.
Snowflake as the context layer. You'll make Snowflake the home for our agents' data—for operational data, and in the longer term for ontological knowledge as well.
Backend integration. Via API and MCP, secured through gateways, and where it makes sense, as a CLI.
Making agents observable. Prompt, tool-use, and cost telemetry end to end, evals as part of the lifecycle, lineage from the user action through the model call to the effect in the core system.
Security as a platform feature. Machine identities and permissions for agents, guardrails against prompt injection, clean secret management, audit trails.

About You

Experience as a software or, ideally, a platform engineer
Hands-on experience with AWS and in data engineering, ideally with Snowflake
Have run LLM-based applications in production, or closely supported doing so
An understanding of agents: tool use, context engineering, evaluation, observability
High agency and the ability to make decisions and explain your reasoning clearly
You use AI daily in your own work

Responsibilities

About the Job

You'll build the path that takes AI into production here:

From prototype to productive application. You build the bridge from "sketched out with Claude in Python" to "running—observed and secured—in operations."
Selecting and integrating components. LLM gateway, container runtime, workflow and agent orchestration, observability, cost tracking. As a team, not alone.
Snowflake as the context layer. You'll make Snowflake the home for our agents' data—for operational data, and in the longer term for ontological knowledge as well.
Backend integration. Via API and MCP, secured through gateways, and where it makes sense, as a CLI.
Making agents observable. Prompt, tool-use, and cost telemetry end to end, evals as part of the lifecycle, lineage from the user action through the model call to the effect in the core system.
Security as a platform feature. Machine identities and permissions for agents, guardrails against prompt injection, clean secret management, audit trails.

About You

Experience as a software or, ideally, a platform engineer
Hands-on experience with AWS and in data engineering, ideally with Snowflake
Have run LLM-based applications in production, or closely supported doing so
An understanding of agents: tool use, context engineering, evaluation, observability
High agency and the ability to make decisions and explain your reasoning clearly
You use AI daily in your own work