Summary

Metrics-driven Senior Site Reliability Engineer (SRE) with 8+ years of experience delivering robust, reproducible systems across diverse technical ecosystems. A polyglot developer with deep expertise in Kubernetes, Terraform and Nix, alongside a broad, adaptable toolset spanning cloud-native architecture, agentic systems, and observability-metrics. I excel at rapidly mastering new technologies at proof-of-concept level to validate architecture early. Focused on finding sustainable fixes to complex architectural bottlenecks, I bridge the gap between rapid technical experimentation and long-term system reliability. Looking to bring a pragmatic, full-spectrum engineering mindset to a forward-thinking development team.

Work Experience

SRE/Cloud Engineer

EvBox | Amsterdam
2022 -
  • Built a Ray-orchestrated, multi-node vLLM inference cluster as a proof of concept (DeepSeek-R1): distributed serving across smaller GPUs, where calls waited for the cluster to scale up before inference began.
  • Owned the Prometheus and VictoriaMetrics observability platform: a central, multi-tenant metrics cluster with HA storage and auto-scaling ingestion, serving as the metrics backbone for 30 microservices.
  • Built the core GitOps platform managing the full lifecycle for 30 microservices across 8+ GKE clusters and 1,000+ nodes, with IAM, DNS, and Vault authentication bootstrapped from scratch in Terraform.
  • Built the company’s HashiCorp Vault deployment from scratch: PKI-based mTLS and KV v2 handling 500+ requests per second, eliminating static credentials across three environments.
  • Designed the Vault-backed custom PKI and Pomerium identity-aware access layer that replaced Google BeyondCorp: certificate-based client validation for every cluster, without its licensing cost.
  • Made SBOM creation, scanning, and lifecycle management a first-class stage of the container image pipeline.
  • Productized CI for 30 product teams: shared GitLab CI/CD pipeline templates and 24 custom base images enforcing Snyk and SonarQube quality gates.
  • Built ephemeral per-branch preview environments and a two-phase GCP project factory with least-privilege custom roles and automatic budget guardrails.
  • Operated production on-call as incident commander in a 5-person SRE team embedded with product teams across the company, on Google’s SRE methodology, and mentored junior engineers along the way.

Site Reliability Engineer

SprintHive | Cape Town
2019 - 2022
  • Automated Prometheus provisioning via custom Terraform modules that dynamically calculated resource limits. Implemented query throttling to prevent Grafana-induced OOM crashes, and defined latency and availability SLOs.
  • Migrated Elasticsearch configuration from manual click-ops to fully codified deployments.
  • Maintained highly available Kubernetes clusters on Google Cloud integrated with AWS storage, and implemented custom Horizontal Pod Autoscaling (HPA) driven by Prometheus latency metrics.
  • Refactored duplicated Terraform infrastructure into dedicated, reusable modules, reducing deployment cycles from days to a single day.
  • Developed a Go-based repository management CLI that automated Git hook installation and Terraform formatting.
  • Ran incident response as incident commander and on-call SRE, authoring blameless post-mortems (Correction of Error documents) after production incidents; participated in Red/Blue team exercises to prepare developers for security incidents.

Acting Software Development Engineer

Amazon Web Services | Remote
2019 - 2019
  • Proposed and prototyped a Machine Learning system to automate abuse case classification by learning from support agent corrections.
  • Developed automated, legally compliant archiving tooling for sensitive customer service content, producing the full audit trails that supported high-stakes legal dispute resolution.

Technical Customer Support Specialist

Amazon Web Services | Cape Town
2016 - 2019
  • Built automation processing 14% of email messages per week, reducing the need for extra staff.
  • Developed Python-based heuristic scanning tools targeting EC2 drive I/O metrics to detect malware installations, alerting customers and preventing unauthorized infrastructure charges.
  • Engineered Python network traffic analysis tooling to identify botnet signatures and malicious communication patterns, producing standardized reports for multiple AWS security teams.
  • Collaborated directly with the EC2 Core Team at AWS Seattle on high-impact security events: DDoS response, abuse operations, and customer-facing security escalation.

IT Support Manager & Field Technician

HX Systems | Somerset West
2013 - 2016
  • Engineered and maintained the physical and network infrastructure for a regional Wireless Service Provider (WISP), deploying multi-site routing topologies and administering core endpoints using OSPF/BGP. Architected the migration from consumer-grade hardware to a highly-available VMware HA hypervisor stack, centralizing critical ISP services, core routing, and authentication on bare-metal virtualization.

Projects

Personal Cloud Platform & Distributed Systems R&D

  • Platform Architecture: Architected and operate a 6-node, highly-available Talos Linux Kubernetes cluster (3 control plane, 3 worker) on bare-metal Proxmox QEMU/KVM hypervisors. The platform orchestrates 210+ pods and 150+ services across 35+ namespaces, governed by strict ArgoCD GitOps for declared state reproducibility.
  • eBPF Networking & Ingress: Designed a Dual-Stack IPv6 three-tier network fabric with FRRouting (FRR) and OSPF peering. Replaced kube-proxy entirely with Cilium eBPF for kernel-level routing and observability, maintaining a production-grade ingress pipeline (MetalLB, 3-replica Traefik, HTTP/3, Gateway API) to evaluate emerging CNCF networking standards.
  • Agentic Telemetry: Operate OpenTelemetry tracing end-to-end: OTel Collector pipelines feeding VictoriaMetrics/VictoriaTraces and Grafana instrument agentic execution loops (token amplification, retry rates): AI systems observable by construction.
  • LLM Platform R&D: Operate model routing and gateway traffic for self-hosted AI platforms (Hermes assistant, Windmill flows), evaluating routing, quota, and failure semantics for LLM-backed services. Nix-based vulnerability scanning and image validation keep the declarative supply chain byte-perfect.

Mnemosyne: Multi-Agent Documentation System

  • Designed and built a LangGraph + LangChain multi-agent system (Planner, Researcher team, Writer clique, Assembler) that generates structured historical reports from an entire corpus of engineering work history.
  • RAG retrieval with HyDE: hypothetical-document embeddings over a ChromaDB vector store, so semantic queries surface the right history even without keyword overlap.
  • Corpus ingestion: the raw git history of an engineering tenure flows through discovery and git-database pipelines into structured, retrievable knowledge.
  • Google GenAI (Gemini API) integration end-to-end; typed, fully tested, and packaged reproducibly with Nix Flakes.

agent-ear: Agentic Voice Interface

  • Open-source agentic CLI for audio and video comprehension: media goes straight to a multimodal model, preserving prosody: the emphasis, tone, and pacing that plain transcription throws away.
  • End-to-end agentic pipeline: the agent validates its own extraction prompts, records within agent-defined constraints, and briefs users through TTS; a versioned commit-enforcement config forces the AI to respect repository standards.
  • Supply-chain and accessibility posture: OpenSSF Scorecard, CodeQL, and vulnix scanning, Dependabot gatekeeping, WCAG-compatible video descriptions; dual-licensed (MIT, Apache-2.0) and installable via Homebrew and Nix Flakes.

Technical Skills

Orchestration & Automation

Kubernetes, Terraform, Nix, ArgoCD GitOps

Networking & Storage

Dual-Stack IPv6, FRRouting (FRR), Cilium eBPF, Ceph CSI

Core Practices

Observability & Metrics, Platform Security (HashiCorp Vault), Agentic Systems