Senior AI Engineer

Nasiko is building the infrastructure enterprises need to run AI agents securely, efficiently, and reliably in production—and we’re looking for a Senior AI Engineer to help build the backend and AI systems that make that real. If you like turning prototypes into dependable services (with observability, policy, and cost controls baked in), this is a rare seat where you’ll own the core plumbing.

The Work

You’ll build and operate the backend + AI infrastructure behind Nasiko’s platform, including:

• TokenOps / FinOps for agents: token usage, cost attribution, forecasting, and optimization systems that help teams understand and control spend.
• Runtime policy enforcement: governed access for agents to models, tools, data, and credentials—so “it works” becomes “it’s safe and compliant.”
• Multi-agent execution: workflow infrastructure for execution engines, retries/recovery, and human approvals when it matters.
• LLM routing + model access: provider-agnostic routing and integration patterns for hosted and customer-managed models/inference endpoints.
• AI + MCP gateways: MCP server/client/tooling and gateway integrations that connect agents to enterprise systems in a controlled way.
• Evaluation + production feedback loops: monitoring, tracing, and evaluation pipelines that improve quality over time—not just in demos.

You’ll also contribute to the platform’s Rust-based services while building AI-facing components in Python, and you’ll own the deployment and operation of what you ship.

What You’ll Need

• Strong production engineering across backend systems (APIs/services/workers), including async/event-driven patterns and clear service design.
• Hands-on Python and Rust experience building and shipping production systems.
• Experience building LLM-powered applications or AI agents that use tool-calling and/or retrieval-style components.
• Production experience operating services with logs/metrics/traces and alerting/observability in place.
• Practical infrastructure skills: Docker, Kubernetes, and CI/CD; plus comfort with databases and queues (e.g., PostgreSQL/Redis and message queues).

Even Better If

• You’ve built LLM routers/model gateways, agent runtimes/orchestration, or inference/model-serving infrastructure.
• You’ve worked on AI security/governance/policy enforcement or AI FinOps/token analytics.

Who Thrives Here

• You enjoy building the “boring but critical” parts—retries, failure handling, structured outputs, and operational visibility—because you know that’s what makes agent systems trustworthy.
• You can move between product intent and system design without losing either: you’ll translate requirements into clean abstractions and then ship them.
• You’re comfortable owning services end-to-end, including deployment and production support, and you don’t wait for someone else to make it reliable.

The Team

Nasiko is a Rust-first company building agent infrastructure for production use. You’ll work across AI infrastructure, backend engineering, and production operations to take agent workflows from prototype to secure, observable services. The team is small enough that your decisions and code will directly shape how the platform works.

Comp, Logistics and Benefits

Location: Palo Alto, Hybrid
Employment Type: Full-time

Nasiko uses Rust for core platform services and Python for AI development, integrations, evaluation, and experimentation.

Non-Traditional Backgrounds

If you’ve built and scaled systems in any environment—startup, enterprise, or open source—and you’ve shipped LLM/agent capabilities into real production constraints, we want to hear from you.