AI Engineer & Solution Architect with a Multidisciplinary Background in Data Science, Front-End Development, and Visual Design

My journey from illustration to programming to AI has given me a unique perspective in transforming complex problems into intelligent, scalable solutions. As an AI Solution Architect & Engineer at Boosha AI, I design and ship production systems that unite logic and creativity to optimize workflows and solve real-world problems.

I build AI agents and multi-agent systems with LangGraph, CrewAI, and MCP, design RAG pipelines with vector databases (FAISS, ChromaDB), and apply MLOps practices with Docker, Kubernetes, and Terraform on AWS and Oracle Cloud. This sits on a solid Machine Learning foundation (Python, PyTorch, TensorFlow, Pandas, NumPy, Scikit-Learn) covering EDA, predictive models, and data visualization. My background in structuring visual information bridges naturally to AI, where logical organization is fundamental.

My work method draws inspiration from productivity and knowledge management practices: I write morning pages to stimulate creativity, use bullet journaling to organize my day, and manage resources with Obsidian, following the Building a Second Brain methodology by Tiago Forte.

When I'm not immersed in data, I'm passionate about TV series, comics, Magic: The Gathering, and Lego—worlds characterized by complex, well-structured systems that mirror my passion for order and logic.

• 🌍 I'm based in Italy
• 🖥️ See my portfolio at veronicaschembri.com
• 🚀 AI Solution Architect & Engineer @ Boosha AI
• 🌱 Currently exploring Kubernetes, advanced Terraform & observability
• ⚡ I'm a Panda, and a super nerd. I love playing Magic: The Gathering.

• Nextcloud Self-Hosted su OCI — Infrastruttura resiliente con Terraform, architettura Pets vs Cattle, monitoring (Prometheus/Grafana), backup automatizzati. REPO

• Veronica Chatbot — Sistema AI full-stack (LangGraph ReAct + WordPress API + React) in produzione su Railway. REPO
• GAIA Benchmark Agent — React Agent con 60% accuracy sul benchmark GAIA, 15+ tool orchestrati con LangGraph.
• AI Interview Prep (CrewAI) — Sistema multi-agente per la preparazione ai colloqui. REPO · Live
• NutriCHOice — Generatore di ricette con LLM + LangGraph (approccio Generate-then-Fix). REPO · Live

• AI Assistant with RAG — Document Q&A + Meal Planner con Retrieval-Augmented Generation. REPO
• Dog Breed Classifier — Deep learning per il riconoscimento razze da foto. REPO

Kubernetes, Terraform avanzato, observability

AI Engineering & Machine Learning

Front End Dev

Tools