Foundations Engineer (Deep Infra)

We work on the systems responsible for:

real-time context ingestion

agent execution and orchestration

reliability for long-horizon AI tasks

low-latency decisioning across distributed systems

If you've worked on systems like:

streaming compute platforms

distributed query engines

real-time OLAP systems

matching engines

large-scale data infrastructure

many of the problems here will feel familiar — but applied to a new category of software.

At Rox, agents are constantly:

retrieving context

making decisions

triggering actions

updating state

The Foundations team builds the infrastructure that makes those loops reliable, fast, and observable.

The Role

We are hiring a Foundations Engineer (Deep Infra) to design and operate the systems that power Rox’s agent runtime.

This role sits at the intersection of:

distributed systems

real-time data infrastructure

agent orchestration

decisioning systems

You will work on infrastructure similar to what powers large-scale data platforms — but optimized for continuous AI decision loops rather than batch analytics.

Many of the systems we are building resemble components from:

streaming compute systems

distributed query engines

large-scale event processing pipelines

matching and routing infrastructure

But with a new constraint: AI agents must operate continuously and reliably in production environments.

What You’ll Do

Design and build core distributed systems powering Rox agents.

Build infrastructure for real-time context ingestion and retrieval.

Develop systems that enable low-latency agent decisioning at scale.

Improve reliability, observability, and fault tolerance across agent infrastructure.

Build execution frameworks that support long-running agent workflows.

Collaborate closely with product engineers and forward deployed engineers to translate infrastructure capabilities into customer impact.

Ship production systems quickly and evolve them based on real-world usage.

What You’ll Bring

Experience building deep infrastructure systems in production.

Examples include:

distributed data platforms

streaming compute systems

real-time analytics infrastructure

distributed query engines

high-scale backend services

Strong intuition for latency, reliability, and scalability tradeoffs.

Experience operating systems where real-world traffic exposes edge cases quickly.

Comfort debugging complex distributed systems in production.

A bias toward shipping systems, learning from production, and improving them continuously.