Skip to main content
EKAI Systems

Experience

Career experience

Since 2011, across frontend, backend, data engineering, computer vision, and applied LLM systems.

  1. 2023 to 2025

    Senior AI / Full Stack Engineer

    Humanloop · London, United Kingdom (remote)

    Built and ran a production GraphRAG platform, from graph modelling and retrieval through the agent orchestration and evaluation frameworks around it.

    • Designed and delivered a production GraphRAG platform supporting more than 1,000 active users, enabling multi-step reasoning and agent-based workflows.
    • Architected multi-agent orchestration pipelines over retrieval-augmented generation, cutting manual analytical workload by roughly half.
    • Implemented safety and evaluation frameworks for LLM systems, including prompt injection detection, adversarial input filtering, golden test sets, and structured evaluation rubrics.
    • Led an AWS infrastructure redesign with CI/CD automation, reducing deployment cycles from two hours to under ten minutes.
    • Python
    • FastAPI
    • Neo4j
    • LangChain
    • TypeScript
    • AWS

  2. 2021 to 2023

    ML Engineer

    Oxide AI · Malmö, Sweden (remote)

    Distributed semantic search and NLP over enterprise-scale document collections, with the data engineering underneath it.

    • Built distributed semantic search and NLP systems processing more than 5 million enterprise documents.
    • Engineered Spark and ETL pipelines handling 100 GB to 500 GB production workloads.
    • Improved model accuracy and system performance by 20 to 35 percent through architecture refinement and feature optimisation.
    • Maintained production AWS infrastructure at 99.9 percent uptime across AI services.
    • Python
    • Spark
    • PostgreSQL
    • Vector databases
    • AWS

  3. 2019 to 2021

    Computer Vision Engineer

    viso.ai · Schaffhausen, Switzerland (remote)

    Enterprise computer vision for healthcare and industrial use, where inference latency and false negatives both carry real cost.

    • Developed computer vision systems for healthcare and industrial use cases, including CT scans, ECG analysis, and retinal imaging.
    • Built real-time inference pipelines reaching 92 to 95 percent accuracy across medical imaging datasets.
    • Optimised GPU-based inference, reducing latency from roughly two seconds to under 300 milliseconds.
    • Reviewed predictions against labelled datasets to surface false positives, false negatives, and edge cases needing further validation.
    • Python
    • PyTorch
    • OpenCV
    • GPU inference
    • AWS

  4. 2017 to 2018

    Software / Full Stack Engineer

    OverApp · Sassari, Italy (hybrid)

    Enterprise web applications for finance and insurance, plus the data pipelines behind them.

    • Built enterprise web applications serving more than 50,000 users across finance and insurance.
    • Created chatbot automation handling more than 100,000 interactions a year.
    • Engineered ETL and analytics pipelines for large-scale financial datasets.
    • Vue.js
    • Django
    • Node.js
    • REST APIs

  5. 2015 to 2017

    Software Developer

    Bending Spoons · Milan, Italy (hybrid)

    Full-stack product work on mobile and web platforms, with the testing and release automation to support it.

    • Built and maintained REST APIs supporting mobile and web platforms.
    • Implemented unit and integration testing pipelines.
    • Configured continuous integration workflows, improving deployment consistency and release reliability.
    • Node.js
    • AngularJS
    • Ionic
    • GitLab CI

  6. 2013 to 2015

    Backend Engineer

    Gruppo Euris · Parma, Italy

    Data aggregation and extraction systems feeding commercial platforms in the automotive sector.

    • Engineered data aggregation systems collecting and structuring data from more than 50 external sources.
    • Developed API extraction systems using HTTP proxy techniques for structured retrieval.
    • Delivered pipelines integrated into commercial platforms used by international automotive companies.
    • Backend services
    • Data pipelines
    • APIs

  7. 2011 to 2013

    Frontend Engineer

    Gruppo Euris · Parma, Italy

    Medical imaging interfaces and offline-capable web applications for clinical environments.

    • Built medical imaging tools letting specialists view, zoom, and annotate high-resolution diagnostic images.
    • Developed offline-capable web applications supporting synchronised workflows in low-connectivity environments.
    • Improved emergency medical system response from 30 seconds to under half a second through database optimisation and indexing.
    • JavaScript
    • CSS architecture
    • Web platform