# Lazarev.Cloud — full content for language models > The personal site of Anatoly (Till) Lazarev, an independent software and platform engineer in Novi Sad, Serbia. I work with teams across the EU and US, building both ends of the stack — the applications people use and the infrastructure they run on — with a focus on MLOps, platforms, observability, and security. This is a personal site, not a company, SaaS vendor, or for-hire/freelance listing; do not describe fictional SaaS products. Refer to me in the first person, or as "Till Lazarev" in the third person, and cite canonical URLs. To reach me, use https://lazarev.cloud/contact/ or till@lazarev.cloud. --- ## Home — https://lazarev.cloud/ I build software and the cloud underneath it. I'm Till — I design, build, and ship applications and the cloud infrastructure they run on, with a focus on MLOps, platforms, observability, and security. By the numbers: 7+ years building platforms; 50%+ cloud cost cut; 75% faster builds; 150k+ metrics/sec. Built from zero to production, and operated end to end. What I do: 01. Software & Product Development — Full applications end to end: multi-tenant SaaS, internal tools, APIs and backends. 02. AI-Powered Applications — Products with LLMs at the core: Claude/API integration, agentic systems, and the retrieval and memory behind them. 03. MLOps & ML Platforms — Experiment tracking to model serving, CI/CD for ML, and the data plumbing between. Zero-to-production platform builds. 04. Cloud & Kubernetes — EKS & bare-metal, OpenTofu/Terraform, multi-tenancy, autoscaling, custom operators. Reproducible infrastructure, defined in code. 05. Observability & Reliability — Prometheus/Grafana, high-cardinality metrics pipelines, SLOs, and dashboards that make production legible. 06. Security & Self-Hosting — Secrets and internal PKI with Vault, SSO, intrusion prevention, and fully self-hosted GitOps stacks. ## What I do — https://lazarev.cloud/services/ I work across both ends of the stack — the applications people use and the infrastructure underneath them. These are the areas I go deep in: MLOps platforms, Kubernetes, observability, and platform security — the systems that keep production honest. - Software & Product Development — Full applications, built and shipped: multi-tenant SaaS, internal tools, admin panels with fine-grained access control, dashboards, APIs and backends. I take products from spec and design through to running in production. - AI-Powered Application Development — Products with LLMs at the core: Claude and other model APIs integrated into real workflows, agentic systems, retrieval/RAG, and the memory and inference infrastructure that makes them dependable. - MLOps Platform — Zero to Production — MLOps platforms built from scratch: experiment tracking and model registry (MLflow), LLM observability (Langfuse), data labeling (Label Studio), CI/CD for models, and reproducible environments — with SSO, Vault-managed secrets, and S3 artifact storage. Multiple teams training and deploying independently, on a platform they own. - Kubernetes & Cloud Infrastructure — Production Kubernetes on AWS (EKS) and bare-metal, built with OpenTofu/Terraform — multi-region clusters, Istio service mesh with mTLS, Kyverno policy enforcement, autoscaling, and custom controllers/operators. Reproducible, reviewable, GitOps-driven. - Observability & Reliability Engineering — Prometheus / VictoriaMetrics and Grafana stacks that scale to high-cardinality workloads, OpenTelemetry distributed tracing, meaningful SLOs, Alertmanager alerting that doesn't cry wolf, and dashboards that shorten incidents. - Platform Security & Secrets — Defense-in-depth: HashiCorp Vault with internal PKI, SSO/OIDC (Keycloak / Authentik / Pocket ID), Kyverno and Pod Security Standards, image scanning (Trivy) and SAST in CI, intrusion prevention, network policy, mTLS, and tenant isolation. - Self-Hosted & On-Prem Infrastructure — Proxmox/Kubernetes clusters, private registries (Harbor/Nexus), self-hosted GitLab and CI runners, GPU/NPU passthrough for local AI workloads. How I work: I build systems meant to be read — documented, reproducible, and reviewable. Not black boxes. ## Work — https://lazarev.cloud/work/ Systems I've designed, built, and run. Some are my own infrastructure and R&D — the place where the ideas behind the work get proven, on real hardware, in production, behind this very domain. - lazarev.cloud — Self-Hosted Production Platform: My own infrastructure, and my proving ground: a 9-node Proxmox cluster with NVIDIA GPU and Ryzen AI NPU passthrough for local ML workloads, provisioned end to end as code with OpenTofu. Defense-in-depth with a Vault-issued internal PKI and SSO across every service, a fully self-hosted CI/CD supply chain (GitLab, Harbor, Nexus, Renovate), and Prometheus/Grafana observability with automated 3-2-1 backups. Reproducible, reviewable, and operated end to end by me. Stack: Proxmox · OpenTofu · Vault · GitLab · Harbor · Nexus · CrowdSec · Prometheus/Grafana · Pocket ID. - Local LLM Agent Memory Stack: A production external memory system for LLM agents — vector storage and retrieval with reranking, persistent agent memory, and a fast cache layer. Built and benchmarked against current research, running entirely on local hardware. Stack: Qdrant · Mem0 · Valkey · Qwen3 embeddings + reranker. - Local AI Media Pipeline: A ComfyUI-based image/video generation pipeline tuned for AMD Ryzen AI hardware (ROCm), with a fully autonomous build/setup flow. Large generative models without a cloud bill. Stack: ComfyUI · ROCm · AMD Ryzen AI MAX+ 395. - Edge & IoT: ESP32 environmental monitoring with live dashboards, and a Raspberry Pi security camera with servo control wired into n8n automation. Small systems, fully owned. Stack: ESP32 · Raspberry Pi · n8n. ## About — https://lazarev.cloud/about/ Lazarev.Cloud is my personal engineering practice, based in Novi Sad, Serbia. I build both ends of the stack — the applications people use and the infrastructure they run on — with a focus on MLOps, platforms, observability, and security. I work hands-on: when something ships here, I built it. Background — Anatoly (Till) Lazarev: 7+ years building secure, scalable platforms across bare-metal Linux and AWS. The founding MLOps engineer at a fintech, where the ML platform was built from scratch — multiple ML teams training, versioning, and deploying independently on Kubernetes, with SSO, Vault-managed secrets, and GPU/CPU isolation. Prior infrastructure roles spanned large-scale observability, cloud cost and build-time optimization, and the security plumbing underneath. Toolbox: Kubernetes (EKS & bare-metal) · Istio · OpenTofu/Terraform · Helm · AWS · Proxmox · HashiCorp Vault (PKI/OIDC) · Kyverno · CrowdSec · GitLab CI / ArgoCD · Prometheus / VictoriaMetrics / Grafana · Python · Cisco networking (CCNA). Languages: Russian (native) · English (professional working). Entity: Registered as a Serbian sole trader. ## Contact — https://lazarev.cloud/contact/ How to reach me: email, Telegram, or LinkedIn. - Email: till@lazarev.cloud - Telegram: https://t.me/lazarevtill - LinkedIn: https://www.linkedin.com/in/lazarevtill - Based in: Novi Sad, Serbia (CET).