Hello, I'm
Hualin Ding
GenAI & agentic AI infrastructure engineer. I build the platforms that let AI agents run reliably in production.
01 about.mdsummary
Cloud Engineer at Workday, focused on designing, shipping and scaling AI agent platforms. I work hands-on from LLM prompting and agent design down to infrastructure-as-code and CI/CD, mostly in Python, on Kubernetes and AWS.
Before AI, I spent years on cloud infrastructure and automation at scale. That background shapes how I treat agents: as production systems that need reliability, observability and governance.
02 agents.mdwhat I build
Team-facing AI agents
Chat-based Q&A, knowledge-base, and infrastructure drift-detection agents.Agent hosting platform
AWS-native infrastructure to host and manage agents across teams.Orchestration layers
Harnesses for reliable, auditable AI workflows.RAG systems
Grounding agents in organizational knowledge.Safe execution
Auth, governance and guardrails for agent actions.03 history.logexperience
Workday
Cloud Engineer- Design and build AI agent platforms, including team-facing Q&A, knowledge-base and infrastructure drift-detection agents.
- Run an AWS-native platform that hosts and manages agents built by different teams.
- Build orchestration layers for reliable, auditable agent workflows, and RAG systems that ground agents in organizational knowledge.
- Define auth, governance and safe-execution models for agent actions.
Juniper Networks
Software Engineer · multi-cloud, Kubernetes, GitOps- Led a 7-person cross-functional team on a zero-downtime AWS-to-GCP migration of critical workloads (compute, containers, databases, storage), cutting cloud cost by about 25%.
- Moved deployments from ECS to GKE: roughly 30% faster deployments and 40% lower resource use through pod-level autoscaling.
- Built GitOps delivery with ArgoCD, Helm and Terraform: deploy time down about 50% and new-environment onboarding about 60% faster.
- Led AWS cost optimization by right-sizing compute and adding storage lifecycle policies: about 35% lower compute and 40% lower storage spend.
- Owned operations of a 15+ TB ScyllaDB deployment, with Grafana monitoring, alerting and automated backups.
Salesforce
Software Engineer · cloud cost, observability- Right-sized 100+ cloud instances used for large-scale ML training: up to 40% better training performance and about 30% lower compute cost.
- Tuned managed relational databases ingesting millions of records a day, adding read replicas and automated backups: about 35% faster queries, 60% faster failover and 25% lower cost.
- Migrated block storage from gp2 to gp3: up to 20% cheaper, with about 30% more IOPS and 40% more throughput.
- Led an observability migration from Splunk to Datadog: over 40% lower logging and monitoring cost, about 50% faster dashboards and alerting, plus tagging standards and alert policies.
QBotix
Software Engineer Intern- Built and maintained Docker images, and automated builds, tests and deploys on CircleCI, cutting manual steps by about 50%.
- Helped migrate an application from on-premises to AWS (EC2, S3, RDS, VPC) with minimal downtime.