SRE · PLATFORM ENGINEERING · LONDON

Lisandro
Mentasti

I've spent the last seven years keeping production systems alive at scale: Kubernetes, on-call rotations, SLOs, among other things. Outside of work, I build AI side projects, spend time with my two kids, and run most mornings. Originally from Argentina, now in London.

Kubernetes Terraform AWS Prometheus Python FastAPI Helm vLLM
Email GitHub LinkedIn Resume

Projects

VIEW ALL →
→ runcast-intelligence/
LIVE ↗

RunCast Intelligence

Semantic search across 4,000+ running podcast episodes. Ask anything, get answers with exact timestamps.

RAG pgvector Whisper
→ winesnap/
LIVE ↗

WineSnap

Photo a wine shelf, get the best bottle. Vision AI reads labels, web search finds prices, LLM picks value.

Vision AI PWA FastAPI
→ inferenceforge/
IN PROGRESS

InferenceForge

GPU inference platform on Kubernetes. OpenAI-compatible gateway with queue, rate limiting, Prometheus metrics.

Kubernetes GPU / vLLM Helm

Writing

ALL POSTS →
2025-02 ml-engineering/ → My Journey into Machine Learning Engineering 3min 2023-02 sre/ → Getting Started with Site Reliability Engineering 5min 2019-08 research/ → Teaching a Neural Network to Hear a Motor Breaking 6min

Now

  • 01 Working as an SRE at Okta, building and maintaining Okta Workflows at enterprise scale.
  • 02 Learning AI infrastructure and building side projects to understand how LLMs actually run in production.
  • 03 Running most mornings.
  • 04 Getting into wine, which led to building WineSnap.

LAST UPDATED MAY 2026 · INSPIRED BY NOWNOWNOW.COM