Senior Software Engineer, Agent Platform (Distributed) Remote

castaigroupinc; United States; Remote; USD 80,000 - 111,000; Full time

Benefits offered by this job

Equity options

Learning budget

Flexible hours

Annual hackathon

Equipment budget

Job summary

castaigroupinc is seeking an engineer to join a high-ownership team building a platform for running fleets of AI agents securely, in cloud or self-hosted environments. The role centers on distributed systems that move agent-based workflows from laptop to production, with end-to-end responsibility across design, implementation, testing and rollout.

You will work in a remote-first, globally distributed team and routinely ship features in one to four weeks.

Qualifications

• Production experience with Go or TypeScript or, if not, strong systems programming skills.
• Experience with at least one cloud platform (AWS, GCP, or Azure) and Kubernetes.
• Strong debugging, performance tuning, and observability skills.
• Fluent written and spoken English.

Responsibilities

• Design and build distributed systems to orchestrate agent fleets at scale across cloud and self-hosted environments with security first.
• Make every agent action observable, auditable, and governable via logging, approvals, and least-privilege controls.
• Develop a remote-first experience so developers and operators can interact with agents from anywhere.
• Collaborate with product and other engineering teams on non-textbook problems.
• Own features end-to-end from design to production rollout in a low-process, high-Agency environment.

Skills

Go

TypeScript

Tools

Kubernetes

Prometheus

Grafana

OpenTelemetry

CI/CD

Job description

castaigroupinc is seeking an engineer to join a high-ownership team building a platform for running fleets of AI agents securely, in cloud or self-hosted environments. The role centers on distributed systems that move agent-based workflows from laptop to production, with end-to-end responsibility across design, implementation, testing and rollout.

You will work in a remote-first, globally distributed team and routinely ship features in one to four weeks.