Model Serving
vLLM deployments with OpenAI-compatible APIs, model lifecycle patterns, startup strategy, batching, cache behavior, and throughput validation.
LLM inference platforms | GPU Kubernetes | Multi-cloud SRE
I build and operate cloud-native AI platforms for vLLM model serving, NVIDIA GPU workloads, API gateways, observability, and reliable data systems across Azure, Google Cloud, and AWS.
Platform Direction
vLLM deployments with OpenAI-compatible APIs, model lifecycle patterns, startup strategy, batching, cache behavior, and throughput validation.
NVIDIA GPU Operator, device plugin, DCGM metrics, dedicated node pools, scheduling controls, quota, and GPU saturation analysis.
Inference routing, authentication, tenant controls, rate limiting, model versioning, long-running request timeouts, and traffic shaping.
Grafana, Prometheus/Mimir, Loki, Tempo, OpenTelemetry, alert rules, dashboards, and runbooks for real operational questions.
Selected Work
Public platform blueprint
Production-oriented blueprint for vLLM deployment on Kubernetes with NVIDIA GPU Operator values, Gateway API routing, Prometheus rules, smoke tests, architecture docs, runbooks, and multi-cloud planning for AKS, GKE, and EKS.
Cloud-native observability
Azure Kubernetes Service platform lab using Terraform, Argo CD app-of-apps, Helm, Grafana Alloy, Grafana Cloud, Loki, Mimir, Tempo, dashboards, scripts, and operational documentation.
Professional profile
Writing and posts around AI, engineering roles, platform work, cloud systems, and how automation changes the practical responsibilities of engineers.
Engineering Stack
vLLM, OpenAI-compatible APIs, model routing, batching, inference gateway design
NVIDIA GPU Operator, device plugin, DCGM Exporter, GPU scheduling, node pools
Azure, Google Cloud, AWS, AKS, GKE, EKS, cloud identity, networking, secrets
Terraform, Helm, Argo CD, GitOps, Bash automation, GitHub workflows
Grafana, Prometheus/Mimir, Loki, Tempo, OpenTelemetry, Grafana Alloy
SQL, Python, APIs, data pipelines, retrieval patterns, operational analytics
Cloud security, secret handling, access control, platform hardening, cybersecurity
Verified badges on Credly across cloud, data, security, and platform operations
Verified Credentials
Total public badges
Certifications
Learning badges
Validations
CKS: Certified Kubernetes Security Specialist
The Linux Foundation
Certification
2023-11-24
2025-11-24
Professional Machine Learning Engineer Certification
Google Cloud
Certification
2023-06-19
2025-06-19
AWS Knowledge: Storage Technologist
Amazon Web Services Training and Certification
Learning
2023-06-15
No expiration
AWS Knowledge: Storage Core
Amazon Web Services Training and Certification
Learning
2023-06-05
No expiration
AWS Knowledge: Architecting (Retired)
Amazon Web Services Training and Certification
Learning
2023-03-25
No expiration
AWS Certified DevOps Engineer – Professional
Amazon Web Services Training and Certification
Certification
2023-03-06
2026-03-06
Business Intelligence Foundation Professional Certification - BIFPC™ !
Certiprof
Certification
2023-02-09
No expiration
Business Model Canvas Essentials Professional Certification - BMCEPC™ !
Certiprof
Certification
2023-01-06
No expiration
Linux and Private Cloud Administration on IBM Power Systems Specialization
Coursera
Validation
2022-12-21
No expiration
AZ-400: Designing and Implementing Microsoft DevOps Solutions
Microsoft
Badge
2022-09-18
No expiration
Microsoft Certified: Security, Compliance, and Identity Fundamentals
Microsoft
Badge
2021-11-13
No expiration