Skylo Technologies - Staff Network Reliability Engineer, Service Assurance AIOps & Automation

• Automation Architecture & Technical Strategy • Own the end-to-end support and development of framework and multi-year roadmap for Skylo’s centralized network automation framework — the single platform that Network Operations (Core, RAN, Cloud, IM NRE teams) and Network Engineering consume for automated diagnostics, remediation, change, and validation. • Own the end-to-end support and development of framework and multi-year roadmap • Serve as the design authority for the automation architecture review: set engineering standards, API contracts, data schemas, and governance guardrails that prevent automation fragmentation across domains and between Operations and Engineering. • Serve as the design authority • Define the platform extensibility model so domain NREs and network engineers compose domain-specific workflows from governed, reusable building blocks — without forking framework code or re-implementing primitives. • Define the platform extensibility model • Own build-vs-buy and tooling decisions (orchestration engine, data lake, observability stack, AI tooling), translating them into capital and run-cost implications for engineering leadership with a clear, defensible rationale. • Own build-vs-buy and tooling decisions • Establish the operating model around the framework: ownership boundaries, readiness gates, promotion criteria, and the cross-functional cadence that keeps Operations and Engineering aligned on one automation backlog rather than competing toolchains. • Establish the operating model around the framework • Centralized Framework & Platform Engineering • Architect the centralized framework: reusable health-check modules, diagnostic libraries, remediation actions, and notification/escalation handlers — composable components with versioned, governed interfaces that domain teams chain into workflows. • Architect the centralized framework • Define the automation API layer: standardized interfaces for alert ingestion (Pub/Sub, Prometheus Alertmanager), action execution (kubectl, GCP API, NF CLIs, network configuration interfaces), and result reporting (ITSM update, Slack, dashboard refresh). • Define the automation API layer • Set codebase and delivery standards: testing frameworks, CI/CD pipelines, linting, unit/integration tests, security scanning, and staged rollout via GitLab CI — and hold the organization to them through reference implementations and architecture review. • Set codebase and delivery standards • DAG-Based Orchestration & Closed-Loop Automation • Define the DAG-based orchestration standard (Apache Airflow or equivalent): every workflow a composable, auditable pipeline of diagnostic steps, decision gates, remediation actions, and validation checks. • Define the DAG-based orchestration standard • Architect closed-loop automation for P3/P4 fault classes across both operations and engineering change: trigger → diagnostic DAG → root-cause classification → remediation → validation → auto-close with a full evidence trail and zero human touch for covered classes. • Architect closed-loop automation for P3/P4 fault classes • Own the progressive automation tier model: Tier 1 (auto-diagnose and recommend), Tier 2 (auto-remediate with human approval), Tier 3 (fully autonomous closed-loop) — with confidence scoring and explicit promotion/demotion criteria based on measured incident outcomes. • Own the progressive automation tier model • AI-Assisted Development & Intelligent Operations • Set the team standard for AI-assisted automation development — Claude Code, Claude Cowork, and GitHub Copilot as workflow-integrated force multipliers — and personally prototype reference implementations to prove the bar: from problem statement to staged prototype in hours, not weeks. • Set the team standard for AI-assisted automation development • Architect AI-powered diagnostic agents and alert-correlation engines that correlate Core NF, RAN, Cloud, and OSS signals into actionable root-cause hypotheses and suppress alarm storms — grounded in trusted data, measurable accuracy, and auditability, not novelty. • Architect AI-powered diagnostic agents and alert-correlation engines • Make AI a structural part of the operating model — how the organization detects, isolates, resolves, reports, and learns — with human-in-the-loop gates, model/agent governance, and explicit thresholds for operational trust. • Make AI a structural part of the operating model • Data Engineering & Operational Intelligence • Architect the real-time data pipelines and the operational data lake (BigQuery or equivalent): structure telemetry, incident records, KPI time-series, and automation execution logs into queryable, governed datasets that enable trend analysis, capacity forecasting, and automation-effectiveness measurement. • Architect the real-time data pipelines and the operational data lake • Set DataOps standards: version-controlled pipeline definitions, automated data-quality checks, schema-evolution management, and pipeline health monitoring — the data layer must be as reliable as the network it observes. • Set DataOps standards • Define the dashboarding and analytics standard (Grafana): high-fidelity operational views for NRE teams, network engineers, Incident Managers, and executive leadership, built from a single governed data layer. • Define the dashboarding and analytics standard • Governance, Security & Operational Readiness • Enforce automation governance: every automated action has a defined rollback path, human-in-the-loop validation for P1/P2 actions, audit logging, and safe-handoff to humans when confidence falls below threshold. • Enforce automation governance • Build security into the operating model: least-privilege identity for automation actors, secrets management, full audit trails on production-changing actions, controlled partner/vendor access, and security exceptions that are explicit, time-bound, owned, and visible. • Build security into the operating model • Account for downstream blast radius: ensure automated actions that touch provisioning, inventory, entitlement, assurance, or reporting are designed without creating OSS/BSS or revenue-assurance operational debt. • Account for downstream blast radius • Organizational Leadership & Cross-Functional Partnership • Act as the automation intake and design authority for all NRE domain teams and Network Engineering: convert surfaced toil and engineering pain into prioritized, measurable automation with defined MTTA/MTTR targets and owners. • Act as the automation intake and design authority • Partner with Ops Platform Engineering on shared event schemas, API contracts, and policy engines; partner with Security, OSS/BSS, and Customer Operations to validate operational impact before rollout. • Partner with Ops Platform Engineering • Mentor and raise the bar for Staff and Senior automation developers; establish the technical-leadership and review pipeline for network automation across all hubs. • Mentor and raise the bar • Drive continuous improvement: track automation coverage, MTTA/MTTR impact per workflow, false-positive rate, human-override frequency, and adoption — every automation is measured, not assumed. • Drive continuous improvement