Karl J. Humma

About Karl J. Humma

Pragmatic R&D Leader with over 19 years of experience scaling SaaS infrastructure and reliability. Currently focused on the transition from traditional DevOps to Agentic Operations.My work centers on building the connective tissue between LLMs and operational data. I lead the development of autonomous system interrogation tools, utilizing the Model Context Protocol (MCP) and Claude Skills to transform how SRE teams perform incident triage and system discovery.Core Focus Areas:Agentic AI: Architecting custom MCP suites to bridge the gap between AI agents and production observability (Splunk, Grafana, Port).Operational Excellence: Driving fiscal efficiency—including 20%+ cloud cost reductions—while maintaining rigorous compliance standards (FedRAMP, SOC2).SRE Leadership: Building \"Team of Teams\" structures that prioritize high-velocity delivery without sacrificing system resilience.I am skeptical of AI hype but a firm believer in AI utility. I focus on building internal tools like the \"OpEx Agent\" to solve real-world engineering bottlenecks through autonomous orchestration.

Experience

1. Operational Excellence, Senior Manager

   Qlik

   Apr 2024 — Present · King of Prussia, PA, US

   Leading the modernization of R&D operations by leveraging AI and advanced architectural patterns.Architected and deployed a standardized, protocol-agnostic Agentic Scaffolding to maximize workflow integration and component reuse via a central.github/copilot-agent.yml manifest.Unified incident investigation and observability across FireHydrant, Port.io, Grafana, and Sentry by building a custom OpEx Agent MCP server.Reduced incident triage time by engineering specialized MCP skills (Incident Commander and ATC) standardized across Claude Desktop and VS Code Copilot Chat.FinOps Implementation: Leading the adoption of the FinOps framework to enhance financial governance, specifically architecting systems for automated cost anomaly detection and root cause analysis.Governance: Automating policy documentation via Generative AI to ensure scalable compliance with reduced administrative overhead.Key Agentic AI Achievements:Developed the \"Incident-Commander\" Claude skill, integrating Splunk, Grafana, Port, and LaunchDarkly via MCP to provide autonomous triage capabilities for global SRE teams.Standardized Agentic workflows within R&D, moving from experimental AI usage to production-ready autonomous reliability patterns, including LaunchDarkly-guarded Kubernetes canary rollouts.

Education

• MCP

  MCP, IT

• Architecting for Amazon Web Services

  Cloud

  2013 — 2013

• Drexel University

  BS, IST

  1993 — 1998

Skills

• Iis
• Dns
• Integration
• Software Development
• Devops
• Agile Project Management
• Operating Systems
• Microsoft Sql Server
• Microsoft Exchange
• Business Discovery
• Architecture
• Qlikview Development
• Help Desk Support
• Professional Services
• Technical Support
• Business Analysis
• Big Data
• Erp
• Agile Methodologies
• Management
• IT Management
• Database Design
• Saas
• Cloud Computing
• Active Directory
• Data Modeling
• Windows Server
• Databases
• Business Analytics
• Information Technology
• Data Integration
• Business Intelligence Tools
• Qlikview
• Continuous Integration
• Business Intelligence
• Testing
• Etl
• Enterprise Software
• Enterprise Architecture
• Network Administration