Principal DevOps Architect

REALTIME SOFTWARE SOLUTIONS LLC • Remote, US,

1d • $158,000 - $190,000 • Remote

About The Position

The Principal DevOps Architect is the senior technical authority for the cloud platform that runs RealTime’s global, multi-tenant clinical-trial services, and drives RealTime’s adoption of infrastructure as code and AI tooling. This is a hands-on individual-contributor architect role rather than a people-management role: you design the platform architecture, set the standards and reference implementations other teams build on, and still write Terraform, build pipelines, and stand up the AI platform yourself. You define how reliability is measured against SLOs, how releases ship, and how the AI/ML platform is built and governed, and you keep the environment HIPAA-compliant and SOC 2 Type 2 audit-ready. You influence products, software, and QA through architecture and example rather than direct authority.

Requirements

• Bachelor’s degree in software engineering or an equivalent combination of technical education and work experience.
• 10+ years in SRE/DevOps/platform engineering delivering CI/CD, REST API deployment, containerization, IaaS/PaaS, data pipelines, and application observability, including time at a senior individual-contributor or architect level (Staff, Principal, or Architect).
• Proven technical authority across teams: you set architecture and standards and influence delivery through expertise and example rather than direct management.
• Demonstrated experience driving adoption of a new practice or platform (infrastructure as code, a CI/CD overhaul, or an AI/ML platform) across multiple teams.
• Hands-on experience with Terraform, including writing reusable modules that other teams consume through self-service, remote state, and changing management in a CI/CD pipeline.
• Experience building and operating CI/CD pipelines for large-scale applications on AWS (GitHub Actions, Jenkins, GitLab, or AWS-native).
• Experience running containerized workloads on Docker and Kubernetes.
• Experience with monitoring and troubleshooting using cloud-native tooling and OpenTelemetry; New Relic experience a plus.
• Linux system administration, Unix scripting, and automation.
• 2+ years in one or more PHP, MySQL, and SQL is a plus, matching RealTime’s stack.
• Experience working in a HIPAA / HITECH / HITRUST / PHI / PII or PCI DSS environment.

Nice To Haves

• Experience operating an AI/ML or GenAI platform in production, including Anthropic Claude via AWS Bedrock or the Anthropic API, or comparable model-serving infrastructure, with cost and guardrail controls.
• LLMOps maturity: evaluation pipelines for non-deterministic output, prompt versioning, token cost attribution, and incident response for AI-specific failures (data modification, unwanted workflow triggering, or exfiltration by an agent).
• Experience supporting LLM applications, agents, or MCP services (for example Claude Code or custom MCP servers), and governing how they access regulated data.
• Familiarity with AI security and governance: OWASP LLM Top 10, key management (BYOK), and alignment to frameworks such as the NIST AI RMF.
• Platform-as-product mindset with a focus on internal developer experience and self-service.
• Experience leading or supporting a SOC 2 Type 2 examination and familiarity with security benchmarks such as CIS, OWASP, PCI DSS, and FedRAMP.
• Experience with policy-as-code, secrets management, and supply-chain security (SBOM, image scanning).
• Experience with FinOps and managing large AWS infrastructure and its cost.
• Experience in clinical research or healthcare technology.
• Self-starter who establishes best practices, ships on time, and communicates complex technical information clearly to any audience.

Responsibilities

• Own the platform architecture and technical roadmap for infrastructure, deployment, observability, and the AI platform, and advise the VP of Software Architecture and engineering leadership.
• Set the engineering standards, patterns, and golden paths for infrastructure as code, CI/CD, and AI tooling, and drive their adoption through reference implementations and architecture reviews.
• Act as hands-on technical authority and mentor to engineers across teams, leading by example in code, infrastructure, and incident response, without direct management responsibility.
• Drive RealTime’s move to infrastructure as code and AI tooling and prevent uncontrolled spread of unvetted AI tools.
• Recommend build/buy decisions for platform tooling and evaluate vendors, with final decisions owned by the VP and CTO.
• Manage all cloud infrastructure as code in Terraform as the single source of truth: reusable modules, remote state, peer-reviewed infrastructure pull requests, drift detection, and automated plan/apply in CI/CD.
• Enforce policy-as-code (for example, OPA or Sentinel) so infrastructure changes meet security and cost guardrails before they merge, with separation between the author and approver of a change.
• Design and deploy AWS infrastructure for performance, availability, recoverability, and security across development, UAT, staging, and production, meeting the CIS Critical Security Controls.
• Maintain and improve the multi-tenant database and hosting architecture in a cloud-hosted environment, including replication and per-tenant isolation.
• Build and operate CI/CD pipelines for large-scale applications on AWS using GitHub Actions or equivalent, with automated build, test, and deployment.
• Own release management and rollback: safe deployment strategies (blue/green, canary), fast and reliable rollbacks, and release gates for QA and customer acceptance.
• Package and run containerized workloads on Docker and Kubernetes (EKS), and automate configuration with tools such as Ansible and Packer.
• Lead the SLI/SLO/SLA program and manage error budgets to balance reliability against delivery speed.
• Operate modern observability using cloud-native tooling and OpenTelemetry (metrics, logs, and distributed traces) to drive down MTTD and MTTR.
• Analyze production events to improve reliability, operability, and customer experience, and lead blameless post-incident reviews.
• Participate in on-call rotations, triage and resolve incidents, and provide workarounds or escalation to service owners.
• Provision and operate the AI/ML platform, including Anthropic Claude models served through AWS Bedrock and, where used, the Anthropic API, along with inference endpoints and RealTime’s internal MCP services, all managed as infrastructure as code.
• Build guardrails and observability for AI workloads: prompt and response logging, rate limits, model and token cost controls, and usage auditing across Claude and any other models in use.
• Enforce the data boundary for AI systems so Protected Health Information is handled only through approved, BAA-covered providers (AWS Bedrock under the AWS BAA, and Anthropic under an Anthropic BAA where the Anthropic API is used), and confirm no PHI leaves the compliant boundary.
• Operate and support the agentic and MCP tooling used across engineering and operations (for example Claude Code and internal MCP services) and govern how that tooling accesses code and data with least-privilege tool access and human-in-the-loop escalation.
• Stand up LLMOps practices: prompt versioning and regression detection, evaluation frameworks for non-deterministic output, token-level cost attribution, and audit logging of agent actions.
• Route inference across a multi-provider model portfolio and manage protocol choices (such as MCP) to avoid vendor lock-in.
• Use AI-assisted tooling to accelerate operations where appropriate, such as incident triage, runbook generation, and log analysis.
• Operate and evidence the platform controls required for SOC 2 Type 2 and HIPAA continuously and support external audits.
• Own secrets management so no credentials, keys, or tokens live in source code, container images, or Terraform state.
• Run supply-chain security for the pipeline: image and dependency scanning, SBOM generation, and vulnerability remediation to defined SLAs.
• Manage cloud cost (FinOps): tagging, budgets, and right-sizing across compute, storage, and AI/Bedrock spend.
• Coordinate with product, development, support, operations, and QA so installation and integration are automated and well-documented.
• Communicate and work effectively across a distributed, multi-time-zone team, and cultivate cross-team collaboration and trust.

Benefits

• The company sponsors health insurance, long-term disability, and life insurance.
• Unlimited Paid Time Off.
• 10 paid Holidays.
• Paid Parental Leave.
• Work Anniversary Bonus.
• Participation in the Employee of the Quarter Program.
• Monthly $100 Connectivity Stipend Reimbursement.
• RealTime matches employee 401K contributions at 100% of the first 3% invested and 50% of the next 2% invested.

What This Job Offers

• Job Type: Full-time
• Career Level: Principal