AI & Agentic Systems
Codex, Claude, and Ollama, including AI subagents built for scoped tasks, agent workflows, local and cloud model orchestration, and secure AI enabled operations.
> whoami
Daniel Ruiz
Senior Enterprise Systems Engineer
I build secure, reliable, and scalable enterprise systems at Reddit. My work spans identity, endpoints, SaaS, security, integrations, and AI enabled automation. After 15+ years in technology, I’ve grown from hands on IT support into enterprise systems engineering. I focus on simplifying operations, automating repetitive work, strengthening security controls, and improving the employee technology experience through practical AI enabled workflows.
- 15+
- years in technology
- ~90%
- manual effort reduced in production by AI automations I created
- 10+
- global office network deployments
Expertise matrix
Systems that enable people and protect the business.
Identity Governance
Identity lifecycle, SSO, MFA, SAML, OAuth, OIDC, and SCIM, with governance and traceable security auditing that supports SOX and other audits with documented results for every approved access request.
SaaS Platforms
Google Workspace, Microsoft 365, Atlassian Cloud, Slack Enterprise Grid, GitHub Enterprise, Zendesk, Asana, 1Password, Oomnitza, Duo, and more, plus custom integrations.
Automation
Python, PowerShell, Tines, Okta Workflows, API integrations wrapped into MCP servers, Atlassian Rovo agents, Codex subagents for scoped tasks, repeatable operations, and AI skills with orchestration guides that give AI clear workflows to follow.
Endpoint & Security
Secure device lifecycle across Jamf, Kandji, Microsoft Intune, Cyberhaven, and CrowdStrike, including managed browsers such as Google Chrome, Brave, and Firefox, application deployment across the device fleet, policy creation, controls, and compliance.
Infrastructure
Deployed and managed identity providers including Cloudflare, Okta, and Authentik.io, with group access controls, Google Cloud service account provisioning, and infrastructure as code using Terraform through peer reviewed GitHub pull requests and controlled change management. Secure, resilient network deployments with identity provider connections that make office network access easier to manage and secure through RADIUS.
/ experience
Professional experience
A career built across enterprise systems, identity, security, automation, infrastructure, and support.
Enterprise platform engineering, identity architecture, automation, infrastructure modernization, and operational leadership for a workforce of 3,000+ employees.
- Leads high-impact infrastructure, identity, automation, and operational initiatives from discovery through production.
- Serves as Incident Commander for high-severity infrastructure and SaaS incidents, coordinating response, stakeholder communication, prioritization, and post-incident improvements.
- Provides technical leadership for IT Support through complex escalations, troubleshooting improvements, visual guides, runbooks, and service-delivery practices.
- Architects enterprise SaaS and identity integrations using Okta, SAML, SCIM, OAuth, RBAC, Google Workspace, SailPoint, Cloudflare, and service accounts.
- Builds Tines, Python, PowerShell, and API automation that reduced manual operational effort by approximately 90% across access management, SaaS governance, network monitoring, and support workflows.
- Leads identity audits, license optimization, service-account governance, 1Password SaaS Manager integrations, and repeatable Terraform deployments for Cloudflare and Okta.
- Directs global network modernization and Meter deployments across 5+ offices in North America, Europe, and APAC, including ISP procurement, site surveys, hardware rollout, and production readiness.
- Architected the migration from pre-shared keys to RADIUS 802.1X authentication, strengthening Zero Trust security and network scalability.
Led identity, endpoint security, and IT operations supporting 250+ employees.
- Managed identity and endpoint security across Okta, Jamf, EDR, NGAV, and DLP platforms.
- Designed SAML and OAuth integrations and automated identity lifecycle workflows in partnership with Engineering.
- Established scalable support operations, onboarding workflows, self-service documentation, and security standards.
- Supported SOC 2 and HIPAA compliance through control validation, governance improvements, and audit readiness.
Owned enterprise identity, productivity, collaboration, and network platforms.
- Administered Okta, Google Workspace, Microsoft 365, Slack, Atlassian Cloud, GitHub Enterprise, Zendesk, and Meraki.
- Designed automated provisioning and deprovisioning workflows that improved security and accelerated onboarding and offboarding.
- Partnered with business leaders to translate organizational needs into scalable technical solutions and platform improvements.
Supported enterprise technology services for 1,000+ employees.
- Improved service delivery through workflow optimization and documentation.
- Designed a zero-touch Mac deployment workflow that standardized provisioning and endpoint lifecycle management.
- Led Zoom Room and collaboration-infrastructure rollouts across global offices.
- Owned Okta SSO integrations and partnered with Finance on budgeting, lifecycle planning, inventory audits, and operational spending.
Operated mission-critical production, streaming, and 10Gb network infrastructure supporting globally broadcast TED events.
- Administered Synology and CrushFTP services and optimized high-speed infrastructure for Video Engineering.
- Designed and developed an internal iOS application integrating with Crestron systems to automate theater operations.
- Improved asset visibility and configuration accuracy through Oomnitza inventory governance.
/ selected impact
Systems I’ve built.
Selected work showing how I take complex systems from architecture through production—balancing security, automation, operability, and the people relying on them.
Company Badge Activation Portal
Live self-service activation with structured exception routing
Built a Tines portal that verifies the requester in Okta, finds or provisions the Brivo user and credential, confirms the badge, and activates it. Guarded fallback paths handle delayed provisioning, ambiguous matches, missing identity data, and failed assignments.
- My role
- Owned the portal, identity resolution, Brivo provisioning logic, user notifications, exception paths, and production release.
- Operational result
- Eliminates manual activation intake while preserving an auditable SOS handoff for unresolved or failed activations.
- Controls & systems
- Codex · Tines · Okta · Brivo · Jira exception routing · normalized outcome records
Webhook trace Liveworkflow.run()Employee request→Okta identity→Tines + Brivo→Badge active
License Reclamation Automation
Production-enabled, safety-first license recovery
Built a webhook-driven workflow for explicit SaaS Manager release decisions. It validates and deduplicates requests, resolves the exact active Okta identity, checks licensing-group eligibility and protected-user conditions, and verifies the post-change state.
- My role
- Designed the control flow, identity and eligibility checks, PowerShell and GraphQL actions, rejection paths, and verification logic.
- Operational result
- Enabled in production to reclaim eligible licenses through a controlled, traceable process with post-removal verification.
- Controls & systems
- Codex · Tines webhooks · Okta · Microsoft 365 · PowerShell · GraphQL · protected-user guardrails
Policy trace Verifiedpolicy.evaluate()Release decision→Eligibility gates→PowerShell + GraphQL→License released
End User Wi-Fi Support Portal
Live self-service triage before support escalation
Created an authenticated portal that correlates the requester in Okta with their signed-in Fleet device, current IP, Meter client, office, and Wi-Fi health. Clean-match requirements and deduplication gates control when a structured support request can be submitted.
- My role
- Owned the portal, cross-system correlation, Meter and Fleet failure handling, usage logging, and controlled SOS ticket submission.
- Operational result
- Front-loads identity, device, location, and network context so employees can self-triage and IT receives a better-qualified escalation.
- Controls & systems
- Codex · Tines · Okta · Fleet · Meter · identity matching · ticket deduplication
Correlation trace Matchedcontext.resolve()Employee session→Okta + Fleet→Meter diagnosis→Support ready
Global Office Network Alerting Automation
Live, location-aware alerting with guided response
Built an end-to-end system that enriches and classifies Meter events, correlates related access point and network-switch symptoms, tracks incident state, and suppresses duplicates before routing targeted Slack alerts to the appropriate local IT team.
- My role
- Owned webhook intake, enrichment, root-cause correlation, location routing, incident state, deduplication, escalation, and recovery handling.
- Operational result
- Turns raw telemetry into one actionable response path, linking responders to the correct Confluence runbook and Jira escalation while preserving human validation.
- Controls & systems
- Tines · Meter GraphQL · Slack · Atlassian Jira · Confluence · severity gates · recovery monitoring
Event trace Routedincident.route()Meter event→Correlate + dedupe→Slack + runbook→Local response
Slack Workspace Custom User Group Syncing
Live preventive access reconciliation
Built a scheduled workflow that retrieves Okta group members, resolves each person to a Slack identity by email, aggregates valid matches, and updates the target custom user group through a controlled synchronization path.
- My role
- Designed the member-processing loop, identity matching, aggregation, count validation, update action, and human-review exception routes.
- Operational result
- Keeps Slack membership aligned with its Okta source while surfacing unmatched identities and suspicious member counts instead of masking them.
- Controls & systems
- Tines · Okta · Slack · email identity matching · count sanity checks · exception notifications
Scheduled trace In syncgroup.sync()Scheduled run→Okta members→Resolve + validate→Slack synced
/ systems
My approach to systems engineering
I do my best work where identity, infrastructure, security, automation, and employee experience overlap — turning ambiguous problems into systems with clear ownership and durable operations.
Secure by design
Identity, access, data protection, and auditability are architectural requirements—not finishing work.
Zero Trust patterns like Cloudflare Access in front of SaaS, Okta as the upstream IdP, and RADIUS 802.1X over shared Wi-Fi secrets.
Automate the repeatable
Automation should remove toil, make outcomes consistent, and leave people with better judgment work.
Production Tines and API workflows that cut manual ops work dramatically while leaving judgment calls and exceptions to people.
Design for operations
Observability, recovery, documentation, and clear escalation paths belong in the initial design.
Network alerting with correlation, runbooks, and local routing; recovery plans that restore capability without bypassing change control.
Make security usable
The strongest control is one people can understand and use without finding a workaround.
Self-service portals for badge, Wi-Fi, and license recovery that verify identity and context before support escalates.
Integrate deliberately
Good integrations reduce fragmentation while preserving ownership, failure visibility, and lifecycle control.
One SSO standard instead of one-off IdP-to-vendor SAML; merge approval stays separate from execution approval and plan-only automation.
Communicate clearly
Complex engineering decisions become valuable when stakeholders understand the tradeoffs and next steps.
Change envelopes with purpose, targets, rollback, and evidence — plus incident command and runbooks stakeholders can follow.
- Architecture through production
- Human-approved change, machine-assisted speed
- Standard patterns over one-off integrations
- Recovery without bypassing control
- Lab work that forces real ownership
/ personal lab
Build to learn.
My HomeLab is where I explore local AI, self-hosted infrastructure, model orchestration, and the operational details that only become visible when you run the system yourself.
Local AI Workflows
I set out to learn how to design my own AI bot end to end—not just call a cloud API, but own inference, memory, voice, and ops on hardware I control. On an NVIDIA Jetson Orin Nano Super, I run local models through Ollama, chat through Open WebUI, and image generation with ComfyUI on the Jetson GPU in low-VRAM mode. From there, I built agents in Python: Viper, a Discord bot with Ollama chat, ComfyUI images, faster-whisper transcription, and SQLite conversation memory; and EVE, a private local agent with a browser UI, response-review boundaries, sqlite-vec semantic memory and local embeddings, MioTTS speech, and a VRM avatar. I use systemd and Docker to keep the services supervised and recoverable, which is where the real learning stuck—fitting multimodal workloads into about 8 GB of shared memory and shipping something that stays up.
- Ollama · Open WebUI · ComfyUI
- Python agents (Viper · EVE)
- systemd · Docker
Jetson AI trace Locallab.infer()Private browser→Agent surfaces→Ollama→ComfyUI · Whisper · MioTTS→Persist + recover
HomeLab Infrastructure
A production-minded home lab built in layers: a UniFi network fabric with separated trust zones, a Server_Core for always-on compute and storage, and a deliberately narrow public edge through Cloudflare Tunnel. ZimaCube runs durable file and service workloads; the Jetson Orin Nano Super owns local AI inference. I design this the same way I design enterprise systems—clear ownership, recoverable services, and intentional exposure.
- UniFi fabric · VLAN zones · dual WAN
- ZimaCube services + storage
- Jetson local AI edge
Layered design, intentional surface Sanitizedlab.network()Dual WAN→Dream Machine SE→Pro Max 24→ZimaCube + Jetson→VLANs + tunnel edge
> contact
Let’s build systems that work better.
I’m interested in enterprise system engineering opportunities where I can own architecture through production across identity, AI-enabled automation, SaaS integrations, endpoints, and infrastructure.
user@darksys.dev:~$ Type help to explore.