Skip to content
Stan

Stanislas Andujar

Platform engineer · DevOps · AI agent builder

Eight years building cloud platforms and CI/CD pipelines at scale. Now shipping AI agents to production — autonomous or leading a team.

Montpellier, France github.com/standujar linkedin x.com/stan_andujar s.andujar@proton.me
Book a 30-min intro Download CV

By the numbers

  • 0 ● live

    merged PRs (lifetime)

    across 50+ repos · refreshed hourly via GitHub API

  • 0 ● live

    PRs reviewed

    public reviews on OSS projects

  • 0 ● live

    stars on contributed repos

    sum of top public repos I contribute to

  • 40M+

    weekly viewers reached

    Bedrock streaming platform

  • 450+

    apps on CI/CD platform

    Enedis · French national grid

  • 1000+

    Kubernetes nodes operated

    Bedrock streaming infra

GitHub sync · (cached)

Now

Wrapping up Soulmates at Eliza Labs.

Soulmates lives entirely in WhatsApp. No app, no profile builder, no swipes. You text an AI agent, it onboards you, it profiles you, it matches you, and it coaches you through the conversation that follows. I led the architecture from day one — LLM-driven pipeline, scale-to-zero agents on GKE, async matching engine, and a custom E2E framework that simulates users against the live agent and grades the run with an AI judge.

The lab is closing. I'm shipping the last mile — and looking for the next platform/staff role where DevOps, backend and AI workloads meet.

Updated April 2026

Why hire me

What I bring, concretely.

  • 01

    Scale-ready out of the box

    Operated production platforms at 40M weekly users, 1000+ K8s nodes, 450+ apps on a CI/CD forge.

    Bedrock · Enedis

  • 02

    Open source operator, not just consumer

    387+ merged PRs across 50+ public repos. 143 PR reviews. 104 PRs merged in elizaOS (18.3k ⭐).

    GitHub · live count

  • 03

    Full-stack platform mind

    I write Go credential rotators, TypeScript backends, Postgres RLS layers, and the YAML that ships them. No silo.

    Bedrock · elizaOS

  • 04

    AI in production, not in notebooks

    Soulmates: scale-to-zero AI agents on GKE, custom conversational E2E with simulated users + AI judges.

    Eliza Labs

  • 05

    Senior IC who multiplies a team

    Tech lead on Soulmates from day one. Owned the Argo Rollouts rollout for 50+ Bedrock product teams.

    Eliza · Bedrock

  • 06

    Cost-conscious by default

    Rebuilt Bedrock load testing platform: ×10 capacity at 90% lower cost. FinOps belongs in platform decisions.

    Bedrock

Career arc

The career arc, in plain language.

The thread: I build the substrate that lets product teams ship faster — and now, that substrate has to handle AI workloads in production.

  1. Technical Lead, Soulmates · Eliza Labs

    Remote

    AI matchmaking on WhatsApp/SMS · no app, no forms — entire experience is a conversation with an LLM-driven agent.

    • Led full technical architecture — every decision from infra to product surface, owned end to end.
    • Designed the entire AI pipeline — conversational onboarding, matching and coaching, fully LLM-driven.
    • Tech lead on a 3-person cross-functional team (product + 2 devs) from day one.
    • Shipped 100+ PRs in 6 weeks while running architecture, spec and team coordination in parallel.
    • Built and deployed full production infrastructure on GKE — scale-to-zero agents, async matching and notification pipeline, complete CI/CD.
    • Built a custom E2E framework from scratch for conversational AI — simulated users via LLM, AI-judge scoring, full pipeline coverage. Made non-deterministic agent behavior reliably testable.
    • TypeScript
    • AI agents
    • GKE / Kubernetes
    • KEDA
    • Terraform
    • PostgreSQL
    • Redis
    • Twilio
    • Stripe
    • Anthropic API

    elizalabs.ai

  2. Full-Stack Engineer, elizaOS Core · Eliza Labs

    Remote · San Francisco

    elizaOS — leading open-source AI agent framework (18k+ stars). elizaOS Cloud — SaaS used by 10,000+ users.

    • Core contributor on elizaOS (18k+ ⭐) — the leading open-source AI agent framework.
    • Architected major Cloud platform work — unified messaging API, centralized server package, serverless/Node unified runtime, multi-tenant Postgres isolation with row-level security.
    • Built and maintained core plugins shipped to the entire ecosystem — Discord, Solana, OpenRouter (streaming), WhatsApp, n8n.
    • Standardized logging, TypeScript build pipeline and test infrastructure across the org.
    • Started as unpaid OSS contributor (Apr 2025), joined as paid team member after a year of consistent shipping.
    • Running both elizaOS Core and Soulmates simultaneously since Feb 2026.
    • TypeScript
    • Node.js
    • PostgreSQL / RLS
    • Docker
    • Plugin architecture
    • LLM integration

    elizaos.ai

  3. DevOps Engineer · Bedrock Streaming

    Lyon · hybrid · freelance

    European streaming platform — ~40M weekly viewers, 1000+ K8s nodes, 50+ product teams.

    • Full Argo Rollouts integration — progressive canary strategies with automated rollback on business metrics (Apdex, error rate, success rate, custom KPIs).
    • Serverless credential rotation system in Go — synced across AWS, Fastly and internal services.
    • Hybrid EC2/ECS load testing platform — ×10 capacity at 90% lower cost.
    • Go API Gateway for multi-cloud Kubernetes pre-scaling via SQS / EventBridge.
    • Infrastructure standardization — Terraform module refactor, GitHub Actions centralization.
    • L3 support across 6+ engineering teams. Open-source contributions to ArgoCD and Argo Rollouts (KEDA, Gloo support).
    • Go
    • AWS ECS / EKS
    • Argo Rollouts
    • ArgoCD
    • Fastly
    • Terraform
    • GitHub Actions

    www.bedrockstreaming.com

  4. DevOps Engineer · EDF

    Lyon · permanent

    EDF — France's largest electricity producer. CI/CD platform serving multiple Java/Angular product teams.

    • Built a full CI/CD platform for multiple Java/Angular product teams.
    • On-demand environment provisioning via Terraform + AWS Lambda.
    • Quality gates with SonarQube, artifact management with Nexus.
    • Observability stack — CloudWatch and Grafana, end-to-end.
    • GitLab CI
    • Terraform
    • AWS Lambda
    • SonarQube
    • Nexus
    • Grafana

  5. Tech Lead DevOps → Chaos Engineer · Klanik · Enedis mission

    Lyon · permanent

    Enedis — France's number one electricity distribution network. Critical national infrastructure.

    • GitLab CI forge supporting 450 applications — 40+ modular templates, autoscaling runners, Vault, Nexus, automated DRP.
    • Industrialized CI/CD across 450+ apps — reusable template library, automated security scans on every pipeline.
    • Chaos engineering program — large-scale GameDays (100+ participants) simulating DDoS, database corruption, exposed secrets on resilient EKS clusters.
    • Kubernetes / EKS
    • GitLab CI
    • Vault
    • ArgoCD
    • Prometheus

  6. DevOps Consultant · Lizeo

    Lyon

    Lizeo — automotive data company. Multi-project Kubernetes operations.

    • Multi-project Kubernetes operations across product teams.
    • Prometheus / Grafana observability rollout.
    • Internal training programs on Docker, GitLab CI and Terraform.
    • Kubernetes
    • Prometheus
    • Grafana
    • Docker
    • Terraform

Case studies

Three deep-dives on real work.

What was broken, what I built, what I’d do differently. Honest, structured, 5–8 minutes each.

Stack & architecture

The stack I run, end to end.

A request hits the top of this stack and travels down. Hover any tile to see exactly where I put it to work.

AI / agents

LLM-driven products in production

Backend & data

Where the application logic lives

Platform & delivery

How code reaches production safely

Infrastructure

The substrate everything runs on

  • AWS · EKS
  • GCP · GKE
  • Kubernetes
  • KEDA
  • Terraform
  • Cloudflare

Projects & organizations

Where my commits land. ● live

These numbers are fetched live from the GitHub API on every visit, cached for one hour.

Organizations

+ 847 private contributions not listed (anonymized by GitHub).

Featured open source

A handful of merged PRs.

Skills

  • Languages

    • TypeScript
    • Go
    • Python
    • Shell

  • Cloud & infra

    • AWS
    • GCP / GKE
    • Kubernetes
    • KEDA
    • Terraform
    • Helm
    • Docker
    • Argo Rollouts

  • CI/CD & DX

    • GitHub Actions
    • GitLab CI
    • ArgoCD
    • Dagger
    • Bun
    • Vitest / E2E

  • Backend & data

    • Node.js
    • PostgreSQL
    • Redis
    • Drizzle ORM
    • REST / API design

  • AI / LLM

    • elizaOS
    • Anthropic API
    • OpenAI API
    • Agent design
    • RAG / memory

  • Frontend

    • React / Next.js
    • TailwindCSS
    • Astro

About

Who am I?

Eight years of platform and DevOps work, two of them embedded with AI engineering teams. I sit at the intersection where infrastructure, application code and AI workloads have to ship together — and where most companies still don't have a clear playbook.

What I'm looking for next

Remote · hybrid · EU. From May 2026.

  • Role

    Tech lead or autonomous senior IC

    Lead a small team (2–4 engineers, juniors welcome) and set the technical direction — or operate as a senior IC trusted to stand up a platform from scratch.

  • Phase

    Build, not run

    Designing for scale, optimizing performance against cost, putting the foundations down. Not pure on-call rotation on something already humming.

  • Mix

    Platform and backend

    Keep writing application code — Go, TypeScript, SQL — alongside the infrastructure that ships it. I don't want to live in YAML only.

  • Ground

    Real scale, real complexity, real AI

    High-traffic platforms, hard distributed-systems problems, or AI workloads in production. The combination is where I add the most value.

  • Setup

    Remote or hybrid in the EU

    Async-friendly culture. Strong written communication. Time zones that overlap with Europe.

Book a 30-minute intro call Email me s.andujar@proton.me

What I believe about the craft

  • A good platform, you forget about it.

    If a product team has to think about the platform to ship, the platform is poorly designed. My job is to make deploying, scaling and observing feel like a reflex, not a quest.

  • A canary judges the change, not the whole system.

    A canary that rolls back because an upstream service is degraded is a canary teams stop trusting. Every signal has to say precisely what it measures and what action it triggers.

  • Observability is also rehearsing the failure before it happens.

    Logs, metrics and traces tell you what is going wrong now. GameDays and chaos drills tell you what could go wrong tomorrow. The Enedis chaos engineering program I ran was less about breaking things than about teaching teams to see issues coming.

  • Rotating secrets belongs on the critical path.

    Given the volume of leaks and AI-generated exploit scripts in the wild today, regularly rotating secrets is no longer an optional best practice. It's a production requirement, baked into the deployment path.

  • A test that teams skip is worth nothing.

    The E2E framework I built for Soulmates is meant to be packaged for any AI product. The real goal — give teams faster feedback without trading off quality. Where it makes sense, let AI generate and refresh the tests so they keep up with the codebase.

  • Cost matters as much as latency.

    Latency, errors and dollars-per-user belong on the same dashboard. The 10× capacity / 90% lower cost rebuild at Bedrock came from treating FinOps as an architecture decision, not a finance ticket.

  • At scale, you accompany teams. You do not impose tools on them.

    Migrating 50+ teams to Argo Rollouts at Bedrock was not a top-down rollout. We started from their pain, understood the friction, then tooled — and let teams take ownership. The senior role is to transmit and accompany, not to do everything yourself.

  • AI in production deserves the same SRE rigor as everything else.

    AI today often still runs on a developer's laptop. The real frontier is bringing what we already do for the rest of the stack — canary, scale-to-zero, multi-tenant isolation, rollback on quality metrics — to LLM-driven systems.

Beyond the keyboard

Based in Montpellier. A few personal projects in open source — DCA automation across Solana and EVM, a Soulmates-adjacent matchmaking bot, custom Claude Code tooling. I read more SRE postmortems than is reasonable. Off-screen: cycling, climbing, and cooking that takes its time.

Contact

Let’s talk.

Best path: a 30-minute call. A detailed email works just as well.

Book a 30-min intro Email me s.andujar@proton.me Download CV