● ndlabs // operator profile

ai engineer
infrastructure

I build end-to-end LLM systems that run in production.

● zsh · interactive
nicolae@ndlabs:~$

type help to explore the system // or scroll to descend

service record

Founding Engineer · AI & Infrastructure

STHO Global

Full-time · Sept 2025 to June 2026 · Abu Dhabi, United Arab Emirates · Hybrid

  • Designed the LLM generation systems at the core of an AI-native platform for live entertainment, including automated venue programming (slots, calendars, budgets) that cut manual booking to a single validation step.
  • Built lead generation for artists and venues, pairing external sourcing (scraping, public APIs, Google Places) with LLM enrichment to surface qualified prospects.
  • Developed reputation analysis that aggregates external review signals into structured, comparable metrics.
  • Engineered the full product around these systems (Next.js, Laravel) and ran the infrastructure (k3s, Terraform, Docker, GitHub Actions) in production.

Software Engineer · AI & Data-Driven Platforms

Pulse Partners

Full-time · Feb 2025 to Sept 2025 · Martigny, Valais, Switzerland · Hybrid

  • Integrated LLM-based assistants into production platforms, backed by a multi-source data pipeline (public APIs, structured datasets) that grounded responses in reliable context.
  • Built the backend services behind the AI: context retrieval, prompt orchestration, conversation memory, with token cost managed as a core constraint at scale.
  • Embedded AI-generated insights into dashboards, removing recurring manual reporting for internal teams.

Full-stack Developer

EarnYourStream

Independent · Oct 2023 to Feb 2025 · Paris, France · Remote

  • Built plugins and servers on the TikTok API enabling real-time interaction between a live stream and a running game, where viewer actions triggered in-game events with no perceptible delay.
  • Engineered the system for the highly variable load of live streaming, running on Kubernetes to maintain the real-time loop under sudden traffic spikes.

Information Security Officer Assistant

Conseil départemental de la Seine-Saint-Denis

Apprenticeship · Oct 2021 to Oct 2023 · Bobigny, France

  • Ran web-application security audits and built the Splunk dashboards used by the team to detect threats early across a large public-sector information system.
  • Reduced incident response time from approximately 6h to 3.5h by tightening SIEM incident-management workflows.

Full-stack Developer

Mayara Clothing

Apprenticeship · Oct 2020 to Oct 2021 · Lyon, France

  • Rebuilt the checkout flow of a live e-commerce store (Shopify, Liquid), including custom payment modules and live stock and order management.
  • Delivered changes directly against a production store handling real orders and customers.

Web Developer / Integrator

OwnLeadr

Apprenticeship · Apr 2020 to Sept 2020 · Lyon, France

  • Optimized front-end performance (Lighthouse scores above 90) in support of SEO and organic traffic.
  • Built and integrated responsive client websites from design hand-off through to production.

selected work

01

little_ansible

Ansible-style configuration management tool in Python, over SSH: copy, template, service, sysctl, and apt modules with a dry-run mode.

PythonSSHinfra
02

kubernetes_cluster_k3s

Lightweight k3s cluster on cloud VMs running WordPress/MariaDB and Hugo, with Traefik ingress and HashiCorp Vault secret injection.

k3sVaultTraefik
03

chroot_manager

Minimal container runtime in Python using chroot and bind mounts, driven by a YAML spec.

Pythoncontainerssystems
04

abstract_vm

Stack-machine bytecode interpreter in C: reads and executes an .avm program on a linked-list stack.

CVMsystems
05

kubernetes_cluster_k8s

Multi-node kubeadm Kubernetes cluster running a 3-tier app (React, Node API, MongoDB), with a private registry, ingress, and RBAC.

kubeadmk8sRBAC
06

pong_js

Pong in vanilla JavaScript on HTML canvas: fixed-timestep loop, single-player AI, and local two-player. Playable in the browser.

JavaScriptcanvasgame
07

flagship // classified

The one build that proves the whole thesis. Recovery in progress.

soon

operator dossier

I build AI products end to end: the LLM systems and data pipelines, the app they power, and the infrastructure that keeps it all running in production.

What I care about most is the unglamorous half of AI engineering. Making generation systems reliable instead of just demo-ready, grounding them in real data, and keeping the infra and token cost under control once real users show up.

French engineer based near Paris, master's in systems and networks. Biased toward systems that are durable rather than clever.

design
LLM-driven systems: generation, enrichment, retrieval, prompt orchestration.
build
the backend and data flows behind them (Next.js, Laravel, Node, Python).
run
the infrastructure myself (k3s, Terraform, Docker, GitHub Actions).
focus
= end-to-end LLM systems
layers
= model · pipelines · app · infra
cloud
= k3s · Terraform · AWS · GCP · Azure
status
= open // freelance and full-time

stack

languages

  • TypeScript
  • JavaScript
  • Python
  • PHP

front / back

  • Next.js
  • React
  • Node
  • NestJS
  • Laravel
  • Django
  • PostgreSQL
  • Redis
  • Supabase
  • Prisma

infra & ops

  • Kubernetes / k3s
  • Docker
  • Terraform
  • GitHub Actions
  • Traefik
  • Cloudflare

monitoring

  • Grafana
  • Prometheus
  • Sentry
  • Splunk

open channel

LET'S BUILD
SOMETHING
THAT RUNS.

if the AI is the hard part, that's the kind of problem I like.

transmit to nicolaedogotaru@gmail.com

githublinkedin
nicolae dogotaru // operator© 2026