01
Engineering Leadership
Building autonomous teams, developing technical leaders, commanding major incidents, and aligning engineering work with business outcomes.
MANAGER OF CLOUD SYSTEMS · THERAPYNOTES
Taylor
Finklea
Leading teams. Building systems. Shipping outcomes.
Leads SRE and DevOps at TherapyNotes: architecting the Kubernetes platform, enabling teams to own its implementation and operations, commanding major incidents, and building the tools and practices behind safe, practical AI adoption.
tip: press ⌘K to jump anywhere
system status ● all operational
Engineering Leadership
teams own outcomes
Platform Engineering & Reliability
reliable production operations
Applied AI Engineering
safe, practical adoption
// 01
Teams, platforms, and applied AI
Engineering leadership, reliable platforms, and applied AI work together to create durable outcomes.
02
Platform Engineering & Reliability
Architecting cloud and Kubernetes platforms, improving delivery systems, and enabling teams to own reliable production operations.
03
Applied AI Engineering
Building AI tools, dashboards, standards, and workflows that make adoption practical, safe, and effective across engineering.
// 02
How I lead
I build teams that own outcomes. That means creating clear responsibility, developing technical leaders, removing processes that do not add value, and staying close enough to the work to provide meaningful technical direction.
Build teams around the work
Formed and staffed an SRE team around strong incident responders, then reshaped responsibilities across engineering groups so ownership better matched each team’s strengths. Developed leads and managers so execution could remain team-owned rather than manager-driven.
Lead calmly through incidents
Serve as Incident Commander during major production events. I bring in the right responders, protect engineers from unnecessary interruption, and keep leadership and customer-facing stakeholders informed while technical teams restore service.
Replace process theater with ownership
Moved teams away from hour-based estimation, time tracking, and manager-controlled sprint mechanics. Put delivery and process improvement into the hands of team leads and engineers, using small experiments and observation instead of sweeping methodology changes.
Stay technical where it creates leverage
Architected the Kubernetes platform and helped establish its initial direction before transitioning implementation and operations to the team. Remain hands-on in applied AI by building tools, dashboards, standards, and practices that help engineering teams adopt AI safely and effectively.
// 03
Selected work
001 Applied AI Engineering Case study planned
A Durable Operating System for AI Coding Agents
Fresh context without losing project state
Created a harness-neutral workflow using structured AI documentation, Ralph loops, tiered model routing, and dependency-aware backlogs so work can move between tools and survive frequent context clears.
002 Applied AI Engineering Case study planned
Hermes Voice
A local coding agent, accessible from an iPhone
Built a SwiftUI client and FastAPI bridge for voice and text conversations with a Mac-hosted CLI agent, including pluggable speech providers and a fully local mode.
003 Platform Engineering & Reliability Case study planned
Open Feelings
Private emotion tracking without accounts, ads, or analytics
Built a free, open-source iOS app for naming and logging emotions, with guided check-ins, on-device storage, private iCloud sync, optional app locking, and portable export.
004 Applied AI Engineering Case study planned
Conductor
An inspectable control plane for verified agent work
Building one explicit orchestration kernel for approval, multi-model dispatch, verification, adversarial review, and durable evidence across a fleet of repositories.
005 Platform Engineering & Reliability Case study planned
Larkline
The line to all your tools
Polishing a fast, extensible Rust terminal command palette with fuzzy search, sandboxed Lua and shell plugins, async execution, and cross-platform distribution.
006 Platform Engineering & Reliability Case study planned
Portview
Native iPhone-to-Mac screen sharing and control
Built an Apple-native remote desktop path without a hosted service, using ScreenCaptureKit, hardware HEVC, encrypted QUIC, Metal rendering, and trackpad-style input.
007 Applied AI Engineering Case study planned
CASE
Controlled Autonomy Safety Engine
Designing a provider- and harness-agnostic safety layer for autonomous agent work: outcome contracts, isolated execution, policy gates, evidence bundles, and outcome verification from external state rather than agent claims.
008 Applied AI Engineering Published
Harness Deck
One review surface across every coding harness
Built a harness-neutral local dashboard where AI coding agents publish durable reports, mockups, approvals, and decisions through a versioned contract.
009 Applied AI Engineering Published
Browser Agents
vision-driven web automation
AI-powered browser agents that navigate and automate complex web workflows: identifying elements by appearance and context instead of brittle selectors, with retries, waits and audit trails.
010 Applied AI Engineering Published
AI Development Environment
AI as an everyday tool
Built the environment and guardrails that made AI a safe, everyday tool for engineering: approved models, local LLMs, evals and observability, and a culture excited to share AI-enabled work.
011 Applied AI Engineering Published
AI Automation Agents
reason · plan · execute
Production AI agents that reason, plan and execute multi-step work: customer comms, lead research/enrichment, email and form automation, and API orchestration. Python-first for reliability.
012 Platform Engineering & Reliability Published
Kubernetes
architect and technical lead
Architect and technical lead for the Kubernetes platform. Evaluated Docker Swarm and Nomad, chose Kubernetes and Helm, and helped establish both managed (AKS) and immutable self-managed (Talos, RKE2 on Micro OS) directions before transitioning implementation and operations to the team.
013 Applied AI Engineering Published
Making AI Safe and Practical Across Engineering
From isolated experiments to everyday engineering practice
Developed standards, tools, dashboards, and a community of practice that helped teams adopt AI safely while preserving engineering judgment and accountability.
014 Engineering Leadership Published
Building an SRE Team Around Operational Ownership
Team structure shaped by the work
Formed and staffed an SRE function around incident response, then realigned responsibilities so ownership matched team strengths and operational needs.
015 Engineering Leadership Published
Process That Earns Its Place
Autonomy inside real guardrails
Evaluated each team practice on the value it delivers, removed time tracking and manager-run sprint mechanics, and grew team autonomy while keeping governance and compliance boundaries intact.
016 Engineering Leadership Published
Leading Major Incidents Without Becoming the Bottleneck
Clear command, protected responders
Serves as Incident Commander during major events: bringing in the right people, protecting technical responders from noise, and keeping stakeholders informed while teams restore service.
017 Platform Engineering & Reliability Published
Zero Trust
smaller attack surface
One of the most complex problems I've owned: Cloudflare Zero Trust and a bastion for access, a hub-and-spoke firewall managed with GitOps, and a move toward ACL-based access with Tailscale.
018 Platform Engineering & Reliability Published
Automation & CI/CD
self-service pipelines
Built CI/CD on Azure DevOps then migrated to GitHub Actions: reusable workflows, OIDC service principals, KeyVault secrets, and Automation Accounts for patching and scripting.
019 Platform Engineering & Reliability Published
Infrastructure as Code
immutable, rebuilt at will
Replaced hand-run PowerShell with Terraform and Packer: custom reusable modules, ephemeral systems, and persistent user data via FSLogix roaming profiles on Windows.
020 Platform Engineering & Reliability Published
Containerize & Monitor
no more 3am pages
Containerized manually-deployed microservices, built a custom CI process with Ansible and Bash, and stood up Prometheus / Grafana / Alertmanager monitoring that was still in use when I left.
021 Platform Engineering & Reliability Published
Migrating to Azure
Over 200 VMs · zero downtime
Planned and executed the migration of two datacenters (over 200 VMs from ESXi to Azure) with zero downtime, plus a cloud-native SIEM to replace legacy on-prem.
// 04
Experience
A decade of growing scope, from frontline support to leading teams, architecting platforms, and building applied AI systems.
Technical Support & IT roles
2011 - 2015Dixie State · ManagerPlus · Woods Cross
Network administration, Linux, and desktop support across Dixie State University, ManagerPlus Solutions, and Woods Cross High School. Frontline IT work that built the troubleshooting and service mindset that shaped the rest of my career.
Support Engineer
2015 - 2017DigiCert
Supported customers through complex certificate and PKI issues, building a strong foundation in secure systems and technical problem solving.
Systems Administrator / PM
2017 - 2019Executech
Managed infrastructure and technical projects across varied client environments, combining hands-on systems work with delivery and stakeholder coordination.
Manager of Cloud Engineering
2019 - 2023Lumifi Cyber
Led cloud engineering through platform modernization: implemented Kubernetes, infrastructure as code, and CI/CD. Built metric observability with Prometheus and Grafana, routed security telemetry to Microsoft Sentinel, and led and implemented security spanning engineering, governance, and regulatory compliance.
Manager of Cloud Systems
2023 - PresentTherapyNotes
Leads SRE and DevOps: architecting the Kubernetes platform, enabling teams to own its implementation and operations, commanding major incidents, and building the tools and practices behind safe, practical AI adoption.
// 05
Capabilities
Engineering Leadership
- People-centric leadership
- Team design
- Team building
- Coaching and 1:1s
- Hiring
- Promotions and career development
- Incident command
- Technical strategy
- Organizational change
- Cross-functional collaboration
- Vendor management
- Cost management
- Lean
- Agile
- SRE
- Kata
- Change management
- Strategic planning
- Conflict resolution
- Active listening
- Project management
- Cloud engineering management
- IT management
AI Platforms and Enterprise Adoption
- Azure AI Foundry
- Google AI Studio
- Google Vertex AI
- Google Agent Development Kit
- Google ADK
- Google Agent Engine
- OpenAI Platform
- OpenAI API
- Anthropic Claude Platform
- Anthropic API
- ChatGPT Enterprise administration
- OpenAI Enterprise management
- Enterprise AI governance
- Production model enablement
- Model quota management
- HITRUST AI readiness
- AI adoption strategy
- AI communities of practice
Agent Systems and Developer Tooling
- Agent orchestration
- Multi-agent systems
- Claude Code
- OpenAI Codex
- Codex CLI
- OpenCode
- Pi coding agent
- Hermes
- Model Context Protocol
- MCP
- Agent Skills
- Claude Code skills
- Codex skills
- OpenCode skills
- Pi extensions
- Ralph loops
- /loops
- beads
- Harness Deck
- Conductor
- Human-in-the-loop workflows
- Multi-model routing
- Context engineering
- Durable agent handoffs
- Structured AI documentation
- Agent verification workflows
- A2A
- AI agent development
- Homebrew
Applied AI Engineering
- Python
- TypeScript
- LangGraph
- Pydantic AI
- Mastra
- LangChain
- Browser automation
- Playwright
- Stagehand
- Skyvern
- Ollama
- LM Studio
- Local LLMs
- LLM evaluation
- AI observability
- LangSmith
- Langfuse
- LangWatch
- Prompt engineering
- Tool calling
- Structured outputs
- OpenAI
- Claude
- Gemini
- ReAct
- Groq
- faster-whisper
- Piper
- ElevenLabs
Platform Engineering and Reliability
- Kubernetes
- AKS
- EKS
- RKE2
- Talos
- K3s
- Rancher
- Helm
- Terraform
- OpenTofu
- Packer
- Ansible
- Docker
- GitHub Actions
- Azure DevOps
- CI/CD
- Spacelift
- Prometheus
- Grafana
- Datadog
- Tailscale
- Infrastructure as code
- DevSecOps
- Incident response
- Observability
- GitHub Codespaces
- Git
Cloud, Security, and Compliance
- Microsoft Azure
- AWS
- Google Cloud Platform
- GCP
- Cloudflare
- Cloudflare Workers
- Microsoft 365
- Entra ID
- Active Directory
- Linux
- Windows Server
- SOC 2 Type II
- ISO 27001
- HITRUST
- HIPAA
- NIST
- OWASP
- Tenable.io
- Nessus
- Microsoft Sentinel
- Microsoft Defender
- Privileged Identity Management
- PIM
- Change Advisory Board
- CAB
- Jamf
- Jamf Protect
- Mimecast
- SecurityScorecard
- IT audit
- Vulnerability scanning
- Penetration testing
- Dependabot
- Kali
- Intune
- Ubiquiti
- Juniper
Software Development
- Rust
- Python
- TypeScript
- JavaScript
- Node.js
- Swift
- Go
- PowerShell
- Bash
- PostgreSQL
- SQLite
- InfluxDB
- MySQL
- SvelteKit
- Tauri
- SwiftUI
- FastAPI
- Vue
- Nuxt
- Supabase
- Firebase
- Cloudflare Workers
- Cloudflare Pages
- Vercel
- Fly.io
- Railway
- Netlify
- Loro CRDT
- FTS
- iCloud
- App locking
- Portable export
- Figma
// WHAT'S NEXT
Let’s build what’s next.
I work at the intersection of principal engineering and senior leadership: building teams that own outcomes, architecting reliable platforms, and making AI practical across engineering. If that’s the leader, architect, or applied AI engineer your organization needs, let’s talk.
BEYOND WORK
- Home lab: Self-hosted infrastructure and Home Assistant.
- Wildlife photography: Birds, insects, and the natural world.
- Birding and naturalism: Exploring the land around my home near Kansas City.
- Daily hiking: Time outside, every day.
