Core reception and reservation flows for hospital products, including reservation API reliability, patient visit logic, and customer-specific data boundaries.
Designed and maintained reception logic from patient visit through clinical workflow handoff.
Hardened reservation API paths through Slack webhook cleanup, taxable/non-taxable fields, query optimization, slow web-reservation POST/PATCH fixes, reservation-count sync locking, reservation-log trigger cleanup, and CRM list polling optimization.
KotlinSpring BootGraphQLSQL ServerPostgreSQL
EKS / Terraform / Observability
Kubernetes Platform Modernization
Helped move a legacy cloud platform toward service-oriented architecture, Kubernetes-based deployment, and measurable production observability.
Supported Elastic Beanstalk to EKS migration work with Terraform, Helm, ArgoCD, Rancher, and Kubernetes operations.
Advanced monitoring through Prometheus, Grafana, Jaeger, Datadog, CloudWatch logs, and KPI dashboards.
EKSTerraformHelmArgoCDIstio
Quant Research / ML Pipeline / Market Data
ML-Trade: Intraday Futures Research Pipeline
Research-first quant pipeline for MES/ES intraday opportunity selection, turning Databento OHLCV and MBP feeds into clean features, labels, deterministic baselines, and model-ready validation gates.
Built the market-data path from raw Databento files into bronze/silver Parquet datasets, strict training matrices, chronological splits, manifests, and reproducible research reports.
Designed the research gate around feature information-coefficient reports, strategy-family benchmarks, horizon and roll-rule sensitivity, lookahead audits, MBP microstructure diagnostics, latency/fill/cost stress, and candidate failure attribution.
Borg Agent Orchestrator: Kubernetes Remediation Research
Research system that turns Borg-inspired workload traces and live Kubernetes telemetry into risk and demand predictors, multi-agent remediation decisions, and dual-cluster comparison evidence.
Built the six-layer path from Prometheus/JSON ingestion and simulator feature extraction to XGBoost risk/demand models, RLlib PPO policy evaluation, Optuna reward tuning, and scoreboard feedback.
Validated the control loop on local Kind clusters with AIOpsLab-style incidents, Prometheus/node-exporter telemetry, HPA/local-Karpenter baselines, live Kubernetes action execution, and repeated-seed policy gates.
