vLLM Metrics in Production
A hands-on guide to vLLM monitoring: the key Prometheus metrics (TTFT, TPOT, queueing, KV cache, swapping), Grafana panels, and alert rules that help you debug latency and plan capacity.
Why vLLM Scales: Paging the KV-Cache for Faster LLM Inference
Why traditional LLM serving wastes GPU memory – and how vLLM’s PagedAttention model enables larger batches, higher throughput, and more predictable latency.
uv: Cargo-like Python Tool That Replaces pipx, pyenv, and more
A concise cheat sheet for uv: managing Python versions, dependencies, virtual environments, scripts, and tools in one fast, cross-platform tool.
Identifying Vulnerable Dependencies In .NET Projects
How to scan NuGet packages for security vulnerabilities using GitLab CI.
Convert Flac to Apple Lossless With FFmpeg
How to use FFmpeg to convert FLAC files to Apple Lossless without losing the original quality and uploading them to Apple Music.
Generating A Lockfile For Python Project Using Github Actions
Using Github Actions and pip-tools to compile a requirements.txt file from your dependencies.
Synchronizing Users From LDAP With Keycloak Using AD Filters
Keycloak allows configuring a custom LDAP user filter for User Federation to select a subset of user entries in Active Directory.
Understand How Services Are Run And Operate In Production
Over the past few years, I've been interviewing dozens of software engineers who didn't know how their developed services run in production.
Choosing Apache Kafka For A New Project – A Questionnaire
You're starting a new project with Apache Kafka. Before setting up broker parameters and writing producers and consumers, what questions should you ask yourself? To ensure a smooth start, I have prepared the following checklist/questionnaire.
My "It's not DNS" story
The story about the DNS resolver, Linux VMs, experienced infrastructure team, and me troubleshooting an incident happened on Sunday morning.