Principal Engineer - Architect (AI Agents)
DevSavant Inc.Talent Solutions company
RemoteSenior
Data & AI
About the role
TL;DR
Architect AI agent platform, ensure robustness, scalability, and observability.
- •We're looking for a Principal Engineer / Architect in AI Agents to own the architecture and technical direction of our AI agent platform end to end.
- •This role sits at the intersection of deep systems engineering and frontier AI.
- •You'll design the agent runtime, own the cloud infrastructure that runs it, build the evaluation harness that measures quality, and set the technical bar for a small but growing team.
- •Key Responsibilities Own the system architecture end to end
- •agent runtime, infrastructure, and delivery.
- •Drive the engineering that makes our agents robust, scalable, and observable enough to run reliably in front of clients.
- •Build and maintain the eval harness that determines whether an agent is actually good.
- •Own how live agents are deployed, monitored, and recovered.
- •Set technical direction: bring proposals, decide what moves next, and drive the engineering roadmap in close partnership with the Founder.
- •Requirements 7+ years of experience in software engineering, systems architecture, or a closely related discipline.
- •Agent-harness mastery is the core of this role.
- •You've built, deployed, and operated agents on real harnesses in production.
- •Fluent across the full agentic stack and you know the failure modes unique to agents.
- •Strong cloud and infrastructure depth
- •you've owned full systems from a blank page and designed for scale, reliability, security, and cost.
- •A real DevOps / SRE backbone: live ownership, incident response, root-cause analysis, and distributed-systems instincts.
- •Strong leadership and judgment
- •you set direction, decompose ambiguity, mentor strong engineers, and communicate clearly enough to be trusted in front of clients.
Required skills
AWSGoKubernetesCI/CDDockerPrometheusGrafanaDatadogNew RelicPagerDutyNginxHAProxyGitLinuxJira
Nice-to-have skills
Go
Domain expertise
ai
Tech stack
AWSGoKubernetesCI/CDDockerPrometheusGrafanaDatadogNew RelicPagerDutyNginxHAProxyGitLinuxJiraConfluencePostmanREST APIgRPCWebSocketsRabbitMQCelerySidekiqNotionSlackAsanaExcelPowerPointGoogle AnalyticsGA4
