• Designed and deployed agentic AI systems to automate internal and customer-facing workflows, leveraging multi-agent architectures and tool-calling frameworks to enable scalable decision-making and execution.

• Built and maintained production-grade LLM-powered services (RAG pipelines, autonomous agents, background workers) using Python, FastAPI, vector databases, and OpenAI-compatible models, ensuring seamless integration with existing platforms.

• Architected secure and resilient cloud infrastructure with strong guarantees around observability, monitoring, and agent failure recovery, implementing API security standards (JWT, OAuth) and model governance practices.

• Partnered with product, compliance, and security teams to align AI systems with regulatory requirements, while mentoring engineers on GenAI architecture patterns and scalable deployment strategies.

• Designed strategic initiatives to enhance decision-making systems for flight scheduling across multidisciplinary teams, primarily engineering and data science. Focused on continuous validation through operational KPIs and direct feedback from end users.

• Led the design, development, and deployment of full-stack products while managing a team of engineers — maintaining a hands-on approach throughout. Tech stack included React, FastAPI, MongoDB, PostgreSQL, Dagster, MLflow, Docker, AWS (ECS/ECR, Fargate, S3, …), and Datadog.

• Drove the strategic implementation of agile methodologies across cross-functional squads, fostering alignment and delivery efficiency throughout the organization.

MonolithAI uses artificial intelligence and data engineering to facilitate and reduce costs in automotive engineering testing for leading car brands in the market.

My main responsibilities are:

• Upgrading libraries and Python versions across different services, as well as refactoring legacy code.

• Extending and maintaining web services to improve the user experience.

• Transforming the infrastructure into multi-platform environments to leverage the benefits of various cloud providers.

• Refactor isolating the DB management, ORM and migrations as a new library

for being exploded by several backends.

• Maintenance and add new features in backend and frontend services.

• Improve user experience following UI/UX research.

• Complete backend and infrastructure development for an internal OCR to

extract information and label thousands of invoices per month.

• Refactorisation of the monolithic project in different autonomous services

including a common async task broker and other integrations.

• Complete dev and production reliable environments.