Apr 2024 – Feb 2026 · 1 yr 10 mos · Menlo Park, California, United States · Hybrid

• Optimizing GPU efficient for ML/AI Models (training/inference) and enabling them to run on different hardware (Nvidia, AMD, Meta proprietary).

Mar 2024 – Present · 2 yrs · California, United States · Remote

• Hana is an engagement engine that uses Voice AI, SMS, and Machine Learning to keep patients connected to their care, improving adherence and outcomes.
• I am working with the CTO and the Engineering team on rapidly scaling their Cloud architecture and AI infrastructure to meet explosive demand from clinics globally; we deploy to AWS and use LLMs to power our AI/Voice models.

Jan 2024 – Apr 2024 · 3 mos · Sunnyvale, California, United States · Remote

• As a contributor to Chalk's open-source offering (https://crashoverride.com/docs/chalk) I have designed and implemented the integration with Backstage (https://backstage.io/) to enrich the App catalog with AppSec and build data.

Apr 2023 – Dec 2023 · 8 mos · Silicon Valley, California, United States · Remote

• Shoreline is a platform that helps engineering teams debug, analyze, and automate away production incidents across thousands of nodes and containers on Kubernetes clusters, and then using AI to generate fully-featured runbooks, at the click of a button, to fully automate the process.
• Marco is providing technical advice helping Shoreline to better identify and address prospective customer pain points, and accelerate their ROI.

Aug 2022 – Feb 2024 · 1 yr 6 mos · San Francisco Bay Area · Hybrid

• At Cruise we are building self-driving vehicles to improve life in our cities, using advanced AI algorithms to ensure parity and, eventually, better-than-human driving behavior.
• Marco is a Principal Engineer with the Infrastructure team, currently building a large-scale PaaS framework to manage all our Cloud resources, based on Kubernetes, and leveraging Crossplane.
• He is also involved on a number of initiatives to ensure our services remain highly available, and can scale up to millions of vehicles, globally, across geographies and markets.

Sep 2021 – Sep 2023 · 2 yrs · Sunnyvale, California, United States · Remote

• Copilot IQ is on a mission to improve quality of life and care for Medicare patients with high blood pressure, heart disease, and diabetes.
• As the first member of the technical team, Marco has been helping the medical team to articulate their technology needs; developed the initial architecture and implementation of the backend systems and AWS deployment automation; and then helped build the Engineering team to its current size.
• Marco is incredibly proud that his initial implementation of the Patient API Server, and the many AWS Lambdas services, now greatly enhanced by a talented Engineering team, are now serving thousands of Medicare patients across the US, and allowing them to reach positive health outcomes and leading a better life than they would have been otherwise able to.

Jun 2021 – Dec 2023 · 2 yrs 6 mos · Portland, Oregon, United States

• BankEvo is a fully integrated financial technology platform enabling program managers, fintechs and market-facing brands to bring financial innovation to life via a direct conduit to community banks, capital markets, payment rails, and other financial market infrastructure.
• Marco is advising BankEvo leadership about technology choices, partner and vendor selection, and critical operational concerns around scalability and operational business continuity.

Apr 2021 – Nov 2022 · 1 yr 7 mos · Silicon Valley, California, United States

• OilChain is innovating in the space of self-flying aircrafts (VTOL) by enabling unassisted landing and take off of helicopters and other aircrafts via a sophisticated combination of advanced hardware (radar, cameras, WiFi) and software (image recognition, blockchain, AI).
• Marco is supporting the company funding efforts, as well as technology selection and implementation.

May 2019 – Apr 2020 · 11 mos

• Adobe Campaign is the clear market leader in Cross-Channel customer engagement messaging, and as part of Adobe Experience Cloud is undergoing a period of exciting growth and transformation, building on the virtually infinite scalability potential of Adobe Experience Platform.
• This was a temporary expatriate assignment, to lead several hundred of talented professionals in the Campaign Engineering and SRE teams, across Europe and North America, infuse new life and adopt DevOps best practices to improve the stability and scalability of the service.
• As our customers are the most recognizable global brands on the planet, we serve hundreds of millions to billions of customer communications, and Adobe Campaign must meet very tight SLAs, while at same time our systems must quickly adapt to rapidly changing network traffic patterns.
• By the end of the assignment, all the stated business and technical goals had been achieved and we could successfully transition the product leadership to our colleagues in Adobe India.

Nov 2018 – Aug 2022 · 3 yrs 9 mos

• Experience Platform, the foundation of Adobe Experience Cloud products, is an open system that transforms all our customers' data —Adobe and non-Adobe — into robust customer profiles that update in real time and uses AI-driven insights to help you to deliver the right experiences across every channel.
• This is a massively scalable, distributed system, deployed in the Public Cloud, serving the most successful and recognizable brands across the globe: we have built is as a collection of composable micro-services, with an underlying event-based architecture and flexible data storage patterns, who allow us to ingest billion of events, for hundreds of millions of users, and yet provide a fast, responsive, user-friendly experience to our customers.
• Reporting to the SVP Engineering, I have overall responsibility to provide Architectural oversight to various new initiatives, working closely with the Architecture Group, and the various project teams, providing guidance around both design and implementation issues, at the same time ensuring operational and cost concerns are properly taken into account.
• Most recently, I have personally re-architected, from the ground up, the design of our Customer Journey Management solution, built atop the Experience Platform capabilities, to be the natural evolution of Adobe Campaign: built as a collection of tens of Java microservices, with an underlying event-driven architecture (Kafka) and deployed on Kubernetes, it is a modern, scalable design which will be able to deliver on very aggressive SLAs, scaling up to tens of billions of email and push messages, and more.

Jan 2016 – Nov 2018 · 2 yrs 10 mos · Cupertino, CA

• At Apple we are continuously improving the architecture, design and deployment of our distributed services as it relates to improved performance, scalability & reliability.
• In addition to driving new technology choices, I was also actively involved in the development and deployment of new back-end services that span across several Data Centers and must provide high availability and low latency in the face of several million request per second (QPS) from 155+ countries.
• Among other projects, I have implemented a high-performance orchestration layer in Python to migrate thousand of workloads to a hybrid Cloud infrastructure (AWS); including the containerization strategy of a legacy monolithic Java application.
• Prior to that, I had design a Data Center replication service (Java, Spring Boot), based on Kafka, using Cassandra as the persistence layer, that had to scale up to the needs of the 1.5BN+ iOS customer base; this has been deployed globally, across AWS and Apple private DCs.
• During my time at Apple, I have had hands-on practice with many open-source technologies: Java, Python, C++; Kafka, Cassandra, MongoDB, Voldemort; Apache Mesos, Kubernetes, AWS CloudFormation, ECS, VPCs; and high availability distributed (NoSQL) stores.

Mar 2015 – Jun 2015 · 3 mos · San Francisco Bay Area

• I have taught the Distributed Computing module to the MS in Analytics students.
• We cover topics such as Hadoop, Spark, distributed systems and the CAP theorem, and several examples of NoSQL databases (MongoDB, HBase and Cassandra).
• This is my first foray into academia and I'm excited about sharing several years' of accumulated knowledge and practice in building distributing systems (and finding out, the hard way, all the weird and wonderful ways in which they fail).

Jan 2015 – Jan 2016 · 1 yr · San Francisco

• In addition to contributions to the Apache Mesos (http://mesos.apache.org) codebase (C++) I have been leading the Core Team to accelerate our ability to serve large-scale installation at Fortune 500 clients, financial institutions and large Enterprise customers.
• This was an awesome opportunity for me to "go back to the shopfloor" and work on a massively scalable distributed technology which I consider both exciting and revolutionary.

Jan 2015 – Jan 2015 · 0 mo · San Francisco Bay Area

• Sourceclear is an early stage startup disrupting the world of software security. https://www.sourceclear.com
• In addition to providing product and strategy advice to the CEO, I have designed and implemented the Platform's search service on Elasticsearch.
• Technologies used: Elasticsearch; RabbitMQ; Spring Framework 4.0; REST APIs.

Feb 2013 – Jan 2015 · 1 yr 11 mos · San Jose, CA

• At RiverMeadow I have been driving the architecture, design and development of our Cloud Migration SaaS: we currently support more than 20 OS variants across Windows and Linux, and all major Cloud platforms (VMware Air/vCloud/vSphere, Amazon AWS, Openstack, Oracle Cloud and Microsoft Azure).
• I personally designed, developed and maintained the external (public) REST API, as well as worked closely with the team to design and implement the internal (private) REST API between the Java frontend and the Python backend (Migration Engine).
• As the VP Eng, my role has been to completely rebuild the Engineering and QA teams, drive the adoption of best practices (rigorous code reviews, testing policies and design-for-production practices) and deliver a robust multi-tenant SaaS platform, suitable for adoption by large Enterprise customers, Cloud Service Providers and OEMs (Cisco, VMware, HP, Oracle, and many others).
• Our software was demoed by both VMware at VMWorld 2014, and Oracle's Larry Ellison at OpenWorld 2014 (http://www.oracle.com/us/corporate/oracle-openworld-news/ellison-tuesday-keynote/index.html)
• We use the standard stack of engineering tools: git (Github), code reviews, Continuous Integration (Chef, Jenkins, TeamCity) and vCloud/AWS for deployment/test - all running under Linux (CentOS and Ubuntu).
• The software stack is composed of Java (Tomcat, Spring, JAX-RS) and Python (Flask, Celery, requests, traits) around an abstract execution model and distributed queues; the persistence layer uses Cassandra and MongoDb.

Oct 2011 – Feb 2013 · 1 yr 4 mos · San Mateo, CA

• My role has been to re-build, motivate and energize the Engineering team, to design and develop a Cloud Enterprise integration platform (iPaaS), enabling large organizations to connect diverse endpoints and execute complex, big data transformations via a "data streaming" paradigm.
• Our system is a truly multi-tenant, fully scalable cloud-based integration engine deployed on Amazon's AWS infrastructure and designed around a RESTful architecture, using a judicious mix of Java, Python, and NoSQL (MongoDB).
• All the internal and external (public) APIs were REST-based and I was personally responsible for the design, development and maintenance of a large subset of them; while working with my team's Tech Leads on the overall design and implementation.
• This was a challenging turnaround exercise, building and motivating a team full of talented individuals; driving the adoption of best Engineering practices around Agile development, code excellence and testing policies (for example, driving test coverage from less than 2% to 70%+); and introducing a rigorous release management cycle and continuous integration practices.
• Working closely with the CTO and the CEO, I was as usual responsible for translating the rapidly changing business requirements into detailed product requirements and provide technical direction and leadership for the Engineering team: I was, ultimately, responsible for the Engineering, Product Development, QA, Operations and Customer Success teams.

Dec 2007 – Oct 2011 · 3 yrs 10 mos · Mountain View, CA

• At Google, I was responsible for the Development and Operational continuity of syndicated services to Google's largest Partners, being accountable for the high availability of a complex set of services ranging from search and ads, mobile applications and YouTube, delivering in excess of $1BN yearly revenues, as well as all the development of internal monitoring and reporting systems and tools.
• Leading Google's Partner Solutions Organization's Engineering team across Europe, US and Australia, my main responsibility was to ensure our development practices were in line with Google's demanding standards, and our release cycles aligned within the larger organization's strategic priorities.
• The role required a deep understanding of Google's massive infrastructure and the technologies that underpin the deployment and operation of online services, with very severe demands on scalability and uptime constraints; being involved in the development of new services and internal tools, it also meant I have become fairly familiar with Google's rigorous software design and development process, as well as designing for scale.
• During my stay at Google, I obtained "code readability" in both Java and C++ and developed my passion for Machine Learning technology.

Jan 2004 – Jan 2007 · 3 yrs · Cambridge, United Kingdom

• We founded Tamarix Consulting to provide management consulting services to Telecommunications operators and other service providers in the Mobile/Wireless technology and network planning sector.
• Profitable from the start, we grew it to generate more than $3M revenues in less than two years along the very basic principle of delivering outstanding value for clients.

Jan 1993 – Jan 1994 · 1 yr · Noordwijk, The Netherlands

• Design and Simulation (C++) of mm-wave (50-60 GHz) ICs for Satellite communications and Earth remote sensing.