Employee Spotlight: Alberto Moran
A Fullstack Engineer at QuadSci tuning application performance and architecting the public APIs that put our predictions inside enterprise systems.
Meet Alberto Moran, a Fullstack Engineer at QuadSci. He works across our Vue.js frontend and Python backend, and is leading the architecture of the public APIs that let enterprise clients consume QuadSci intelligence directly. Here's what brought him here and what he's building now.
Why QuadSci?
QuadSci bridges the gap between raw telemetry data and actionable predictive intelligence. Most customer platforms rely on subjective surveys; QuadSci processes real behavioral signals at scale to predict churn and expansion. Joining QuadSci directly aligned with my personal engineering goal: building high-performance systems where machine intelligence isn't an isolated experiment, but is tightly integrated into a resilient, enterprise-grade architecture.
What's your background?
I am a Fullstack Engineer with a strong backend orientation. My background centers on high-concurrency API design, search/vector engines, and scalable systems.
Core Stack: Python (FastAPI) and Java (Spring Boot) on the backend; TypeScript and Vue.js on the frontend.
Data & Search: Deep experience with relational databases (PostgreSQL), document stores (MongoDB), and specialized retrieval engines (Solr, Milvus).
Architecture: Containerized microservices designed for low latency, clean state management, and minimal memory overhead.
My fingerprint: backend-driven fullstack engineering. I bridge complex server-side data processing with clean, reactive client applications.
What are you building right now?
Right now, my work focuses on two core fronts.
Fullstack performance and tuning: I fine-tune QuadSci's application stack — reducing render cycles and state bloat on our Vue.js frontend while optimizing our Python FastAPI backend for low-latency request handling.
The Integrations epic: I am leading the architecture of our major Integrations epic. Previously, our APIs were designed strictly to serve our internal frontend. I am building public-facing backend APIs designed to be consumed directly by enterprise clients and external platforms.
The technical challenge: transitioning from internal UI endpoints to public enterprise APIs requires shifting focus to strict schema versioning, granular rate-limiting, robust authentication, and deterministic contracts. The intelligence QuadSci generates is only as good as its accessibility, and these APIs enable clients to embed our predictions directly into their own systems.
What does accurate and useful intelligence look like to you?
A model output that "technically works" just returns a 200 OK with valid JSON. Useful intelligence satisfies three principles.
High-density signal: delivering precise data over a wall of raw telemetry noise.
Traceability: every prediction must be grounded in verified behavioral data, acknowledging boundaries when confidence is low.
Seamless ingestion: payload structures so clean that downstream client pipelines can consume them without secondary transformation.
One thing people should know about me
Build for operational clarity first, cleverness second.
Code is read far more often than it is written. Whether I'm optimizing a complex query, tuning a Vue store, or designing a FastAPI endpoint, I structure systems so they can be easily maintained under pressure. I bring that same structured mindset outside of work — from tracking performance metrics during high-endurance workouts to building complex engineering assembly kits.