on AWS Step Functions and large-scale serverless systems
Hong Kun Tian
Backend / Platform Engineer · Vancouver, BC
Open to backend and platform engineering roles in New York City, Seattle, Vancouver, Montreal, and Toronto.
Backend and distributed-systems engineer building reliable workflow platforms and developer tooling.
Backend systems, workflow platforms, and developer tooling.
I work on AWS Step Functions, owning production reliability, workflow-platform capabilities, global region launches, deployment automation, and operational tooling.
Current focus
production infrastructure, monitoring, and rollout safety
LLM agents and human-reviewed operational automation
Focus areas
Distributed systemsWorkflow orchestrationBackend servicesService reliabilityInfrastructure automationDeveloper toolingLLM agents
About
Production systems work backed by measurable reliability gains.
I work primarily on distributed systems and backend platforms. At AWS Step Functions, I have built customer-facing workflow capabilities, modernized production infrastructure, supported launches across global AWS Regions, and created operational tooling that makes service changes safer to deploy and easier to review.
Education
B.Sc. Honours Computer Science, McGill University - GPA 4.00/4.00; Alan Shaver Scholarship
Experience
4+ years of backend and reliability work on AWS Step Functions.
Primary role
Software Development Engineer II (L5), AWS Step Functions
Amazon Web Services · Vancouver, BC ·Apr 2024 - Present
Owning production reliability, workflow-platform capabilities, global region launches, deployment automation, and operational tooling for AWS Step Functions.
Highlights
- Owned end-to-end modernization of Step Functions alarm infrastructure across roughly 40 AWS Regions, consolidating 1,016 pipeline deployment groups to 75, cutting package build time from 93 to 26 minutes, and aggregating duplicate paging.
- Built an SOP-driven LLM agent that reduced manual quota-safety triage by approximately 80% across 10-20 weekly requests while retaining human review.
- Owned Step Functions work from internal availability to GA in an isolated AWS environment, diagnosing cross-service failures through remote operators and launching in September 2025.
Earlier experience
Earlier roles built the same foundations that still show up in the current work: UI systems, automation, APIs, and testing.
Amazon Web Services
Software Development Engineer (L4), AWS Step Functions · Jul 2022 - Apr 2024
Helped deliver state machine versioning and aliases, and built throttling, monitoring canaries, and global rollout infrastructure for the ValidateStateMachineDefinition API.
Amazon Web Services
Software Development Engineer Intern, AWS Step Functions · May 2021 - Aug 2021
Built reusable React components, workflow visualizations, and cross-service integration tests for AWS Step Functions console integrations.
Ericsson
Software Developer Intern · Jan 2021 - Apr 2021
Automated Jenkins deployment workflows across VMware and OpenStack environments using Bash and Groovy.
UMAknow
Software Developer Intern · Jun 2019 - Aug 2020
Built cloud-infrastructure diagram generation, a server-side REST API, responsive UI, and automated acceptance tests for Cloudockit.
Selected Work
Recent systems work and internal tooling.
Developer-facing workflow capabilities and validation.
Workflow platform features
Contributed backend capabilities for state machine versioning and aliases, workflow validation, and safer production rollouts.
- Helped deliver state machine versioning and aliases by implementing publish-path idempotency, legacy compatibility, CloudFormation resource behavior, and integration tests.
- Built throttling, monitoring canaries, and global rollout infrastructure for the ValidateStateMachineDefinition API.
AWS Step FunctionsBackend servicesJSONataCloudFormationIntegration testing
Infrastructure modernization and safe launches across global AWS Regions.
Global rollout and production reliability
Owned reliability work spanning alarm infrastructure across roughly 40 AWS Regions, deployment automation, an isolated AWS launch, and operational quota tooling for Step Functions.
- Consolidated 1,016 pipeline deployment groups to 75 and cut package build time from 93 to 26 minutes.
- Owned Step Functions work from internal availability to GA in an isolated AWS environment, launching in September 2025.
AWS RegionsCI/CDDeployment automationCloudWatchService reliability
Human-reviewed automation for complex quota-safety triage.
SOP-driven LLM automation
Built an LLM agent with reusable tools that applies a complex quota-safety decision tree while keeping engineers in the review loop.
- Reduced manual triage by approximately 80% across 10-20 weekly requests.
- Encoded an SOP-driven decision tree through reusable agent tools.
LLM agentsTool callingSOP-driven automationHuman-in-the-loopOperational tooling
Contact
Open to backend and platform roles with meaningful systems ownership.
Open to backend and platform engineering roles in New York City, Seattle, Vancouver, Montreal, and Toronto.