$whoami
Darío Fernández
backend engineer @ freepikSeven years building the platform behind one of the world's largest creative ecosystems — from monolith to microservices, from cron jobs to AI-powered pipelines.
$git log --reverse --since=2019
The journey
-
2019 – 2022
Platform & identity
Joined the team behind Freepik.com and its internal back-office tools. Built and maintained the company-wide SSO / OAuth platform and the login anti-abuse layer — IP intelligence and device fingerprinting protecting sign-in across every brand.
PHPVueRedisOAuth2 -
2023 – 2024
Core APIs & contracts
Moved to the heart of the platform: the API layer serving photos, vectors, icons and AI-generated images to web, mobile apps and partners. Hardened webhook security with HMAC signing, tuned multi-brand caching, and became long-term owner of the OpenAPI specifications and shared domain events that keep dozens of services speaking the same language.
PHPOpenAPIDDDRedis -
2025
Video & content pipelines
Designed and evolved an event-driven pipeline that stages, validates and enriches video and templates at scale — cloud messaging, storage and AI-assisted analysis. Closed real-world attack vectors (zip bombs, symlink reads) and shipped anti-abuse rate limiting for downloads.
PHPPythonPub/SubSecurity -
2025 – 2026
Community, AI & product
Built a new generation of async Python gRPC microservices — hierarchical comments and reactions with bulk operations and multi-language SDKs — and integrated them into the profile platform. Internationalized LLM-generated related searches into dozens of locales, and shipped product features across the creative editor and a modern TypeScript / React monorepo.
PythongRPCLLMTypeScript
$tree ~/impact/
Areas of impact
impact/
core-apis/Core APIs & platform
High-traffic APIs serving creative assets to millions of users across Freepik and Flaticon — caching, reliability and speed at scale.
identity/Identity & security
Company-wide SSO / OAuth, login anti-abuse with IP intelligence and device fingerprinting, signed webhooks, vulnerability mitigation.
community/Community & social
Hierarchical comments and reactions as async Python gRPC microservices: clean architecture, bulk operations, SDKs for every consumer.
video-pipeline/Video pipeline
Event-driven ingestion, validation and derivative generation for video and templates, enriched with AI-assisted analysis.
search-ai/Search & AI
LLM-generated related searches: rollout to dozens of locales and output hardening against model hallucinations.
contracts/API contracts & data governance
Long-term stewardship of OpenAPI specs, CI-validated data contracts, domain events and the shared development data model.
$git log --grep="Co-Authored-By"
AI-augmented engineering
I didn't just adopt AI tools — I rebuilt the way I work around them. Everything below is from the last six months alone.
# shipped to 15 production repositories
- Spec-first: every feature starts as a written design an agent can execute
- Test-driven development, with AI writing and running the loop
- AI-assisted code review before any human sees the diff
- Parallel agents for exploration, migrations and audits
- Persistent memory, so agents keep context across sessions
- Shared agent tooling and context, built with and for my team
$ls ~/stack/
Tools of the trade
$./contact.sh
Let's talk
Open to conversations about backend architecture, platform engineering and AI-augmented workflows.
$