Agent 01Code
Build, test, and ship.
Cloud Automation · AI Infrastructure · Agent Systems
Frank Zhang
I build AI infrastructure, cloud automation, and agent workflows with orchestration, memory, and tool use.
Cloud automationCloud environments, automation, and operational tooling.
Agent systemsReasoning, orchestration, memory, and tool use.
AI infrastructureDeployment, monitoring, and recovery for AI workloads.
Scroll to explore
02 / EVOLUTION
From infrastructure
to agent systems.
My experience covers mainframe support, infrastructure, cloud engineering and automation. I now apply that background to agent systems.
Engineering focus: repeatable workflows, monitoring, and operational controls.
2017MainframeRBC · systems support
2018InfrastructureTTC / Canadian Tire
2019CloudTD · cloud hosting
2020 →AutomationGeotab · cloud platform
NowAI SystemsAgents · memory · LLMOps
03 / ORCHESTRATE
Agent
orchestration.
I work on agent workflows for planning, research, coding and automation, with task delegation, monitoring and human review.
Agent Orchestrationplan · delegate · observe · improve
Agent 02Research
Read, compare, extract signal.
Agent 03Experiment
Run experiments and review results.
Agent 04Observe
Watch systems and surface drift.
Agent 05Automate
Automate repeatable tasks.
Human layerCuriosity
Choose the task and review the result.
04 / FEATURED PROJECT
Agent
Memory Bridge.
AMB stores explicit project decisions for later recall, with their sources and revision history.
- 01 / DecideRecord the decision and its reason.Record a reviewed decision with its context and source.
- 02 / CheckRecall selectively.On a later task, relevance, revisions, and governance shape what is returned.
- 03 / ContinueCheck against current evidence.Use selected memory as a starting point, then check it against live evidence.
Conceptual workflow · not a live activity feed or automatic record of every session.
05 / SELECTED SYSTEMS
Selected
engineering work.
Cloud automation, agent infrastructure and research tooling. The entries below describe the work, methods and available sources.
Developer systems
Automation & platform engineering
Cloud automation and platform engineering at Geotab, using GCP, Kubernetes, Python and CI/CD.
Context
- Work
- Built and operated automation around cloud infrastructure and developer workflows.
- Tools
- GCP, Kubernetes, Python and CI/CD.
Agent infrastructure
Long-running agent workflows
Multi-session agent experiments in orchestration, memory, evaluation, tool use and recovery.
System evidence
- Problem
- Project decisions and their reasons can become scattered across coding sessions.
- System
- AMB stores explicit decisions for later recall, with provenance and revision boundaries.
- Boundary
- Durable decisions stay distinct from current repository facts; AMB is not an automatic transcript archive.
- Evidence
- The public repository documents runnable evaluation checks and their limits; these are not broad productivity results.
Sources: AMB documentation; evaluation scope.
Research automation
Research agents & evaluation
Research agents and evaluation harnesses using data pipelines, retrieval, classification and measurable feedback.
Context
- Focus
- Research agents and evaluation harnesses.
- Approach
- Data pipelines, retrieval, classification and measurable feedback.
06 / EXPERIENCE
Engineering
experience.
My roles have covered mainframe support, infrastructure, cloud hosting, automation and platform engineering.
Dec 2025 — now
Senior Cloud Automation
Geotab
Cloud automation, platform engineering, and increasingly AI infrastructure / agent-oriented developer tooling.
Apr 2020 — Nov 2025
Cloud Automation Engineering
Geotab
Built and operated automation around cloud infrastructure and developer workflows.
Jul 2019 — Apr 2020
Cloud Engineering Associate
TD
Cloud hosting and engineering, after earlier infrastructure and operations internships.
2017 — 2018
Infrastructure & systems roles
RBC · TTC · Canadian Tire
Mainframe support, infrastructure procedure and control, cloud operations, automation, and release management.
07 / BUILDING BLOCKS
Tools
and methods.
Cloud infrastructure
GCP, Kubernetes, deployments and cloud operations.
Platforms & runtime
Developer and agent environments, runtime state and failure handling.
Automation
Workflow automation with checkpoints, tests, monitoring and rollback.
Evaluation & observability
Checks for correctness, drift, context quality, latency, cost and failure modes.
08 / CONNECT
Get in touch
about a project.
I am a senior cloud automation engineer in the Greater Toronto Area. My interests include AI infrastructure, agent systems and developer platforms.
Emailsend message
GitHubcode & projects ↗
LinkedInexperience ↗
Back to topreturn to top ↑
Frank ZhangGreater Toronto Area · Cloud Automation / AI Infrastructure / Agent Systems