Principal Agentic AI Platform Engineer

Avaya

3d•$179,127 - $234,100•Remote

About The Position

Avaya is seeking a hands-on, forward-thinking Principal Platform Engineer to pioneer the integration of Agentic AI into its core engineering ecosystem. The role designs, builds, and scales reusable agentic systems that transform platform development and day-two operations. It extends cloud-native practices with secure agent tooling and rigorous evaluation frameworks to establish a scalable blueprint for autonomous operations across the organization, with a strong focus on security, compliance, and systems reliability. The engineer will establish shared agentic platform patterns and embed AI-supported workflows into real platform delivery. They will work with platform, SRE/CloudOps, and security teams to determine where agents may observe, recommend, or act, and will coach engineers in adopting these capabilities. Success is measured through delivery and operational outcomes such as cycle time, toil reduction, diagnostic lead time, quality, safe automation, and reliability. Applicants must be authorized to work in the United States without visa sponsorship now or in the future. The listed pay range is $179,127 to $234,100 plus benefits; actual total compensation depends on qualifications, skills, and competencies.

Requirements

At least 7 years of experience in software/platform engineering, DevOps, or SRE, with production Kubernetes, Terraform, and GitOps experience.
Strong Python or Go software engineering, API design, testing, integration, and debugging skills.
Hands-on experience building agentic workflows with LLM APIs, tool/function calling, and orchestration; ability to explain working systems, evaluations, and failure cases.
Working understanding of Azure, GCP, or AWS; identity; Kubernetes APIs; Argo CD; CI/CD; observability; and change processes.
Practical knowledge of Kafka, Redis, relational and document databases, API gateways, and Kubernetes networking and security, including the ability to reason about scaling, deployment failures, observability, and day-two recovery.
Experience integrating with Prometheus/Grafana, Loki, Thanos, Alloy, or comparable metrics, logging, and telemetry backends to build reliable diagnostic workflows.
Strong security judgment concerning privileged tools, least privilege, zero trust, and data access in regulated enterprise environments.
Ability to coach technical peers, standardize patterns, and introduce practices across teams without sacrificing operational reliability.
Comfort working at a startup-like pace with ambiguous requirements and rapid iterations, and delivering measurable engineering outcomes rather than demonstrations.
Bachelor's degree or equivalent experience in Computer Science or Engineering.

Nice To Haves

Experience with MCP and agent-tool protocols, retrieval where useful, event-driven systems, policy engines, sandboxed execution, and human-in-the-loop workflows.
Experience with FedRAMP, PCI DSS, or SOC 2 technical evidence requirements and auditable automation.
Experience with cloud networking and platform services such as Istio, Keycloak, cert-manager, API gateways, and DNS.
Experience building or operating agents for diagnosis, remediation, deployment safety, or infrastructure change at scale.

Responsibilities

Create shared agentic platform engineering patterns, including tool interfaces, reusable skills and workflows, coding conventions, evaluation harnesses, telemetry, and adoption practices.
Develop agents and tools integrated with Terraform, Kubernetes, Argo CD/GitOps, cloud APIs, CI/CD, API gateways, Kafka, Redis, databases, observability systems, and change-management systems.
Use AI to accelerate platform development through module and chart creation, code review, infrastructure-plan interpretation, test generation, dependency analysis, and safe pull-request workflows.
Engineer day-two agents for health and drift detection, incident triage, correlated diagnostics, root-cause hypotheses, capacity analysis, upgrade readiness, runbook assistance, and recovery verification.
Build agentic development workflows for Kafka, Redis, relational and document databases, and API gateways, including deployment, scaling, configuration checks, upgrade planning, performance diagnosis, capacity assessment, and secure changes.
Use operational signal sources such as Prometheus, Grafana, Loki, Thanos, Grafana Alloy, and OpenTelemetry to create safe, correlated, retention-aware telemetry retrieval and verifiable diagnostic recommendations.
Develop controlled operational workflows that progress from read-only diagnosis to recommendations, gated changes, and policy-authorized execution where justified.
Enforce security and operational safeguards, including scoped identity, least privilege, environment isolation, secrets protection, prompt- and tool-injection resistance, input/output validation, human approvals, audit trails, timeouts, budgets, kill switches, and rollback boundaries.
Evaluate agent correctness, false positives, tool behavior, failure handling, security conformance, and outcome quality before production rollout.
Embed compliance requirements in agent workflows, including explainable decisions, traceable evidence, approved change paths, data minimization, and continuous control validation for FedRAMP, PCI DSS, and SOC 2 environments.
Partner with security, SRE/CloudOps, and platform teams to define agent permissions and responsibilities, and document accountable owners and escalation paths.
Coach engineers through pairing, playbooks, examples, and metrics to make agentic development a team capability.
Lead adoption by embedding agentic development into platform delivery, shipping working patterns quickly, and improving them through measurement and feedback.
Measure adoption and value through cycle time, manual toil removed, diagnostic lead time, quality, safe automation rate, and reliability.