Marcelo
Mottalli

Staff Software Engineer & Computer Vision Researcher. I solve complex problems at scale — specializing in low-level optimization, distributed systems, and AI. Currently building Shopify Payments in Dublin, Ireland.

Marcelo Mottalli

Experience

Two decades building high-performance systems — from e-commerce platforms to biometric research.

Shopify
Staff Software Engineer

At Shopify, a leading e-commerce platform, I work as a Staff Software Engineer in Shopify Payments, our main revenue source processing $41.1B in 2025 transactions. As a Staff Engineer, I collaborate with other technical teams to make key technical decisions for the product.

  • Led a project to revamp the refund system, preventing fraud-related losses and saving $2M annually through improved security measures.
  • Working on integrating AI into dispute resolution to enhance how we manage customer disputes.
  • Ruby on Rails, MySQL, Yugabyte, BigQuery, LLMs (OpenAI)
Zendesk
Staff Software Engineer

At Zendesk, I worked on the team responsible for Zendesk Explore, our analytics product. The team implemented, executed, and monitored ETL pipelines for analyzing customer data at large scale, primarily using Hadoop (Scala) in AWS.

  • As a Staff Engineer, led the design and implementation of a generic data ingestion system using Java and Reactive Vert.x, providing a unified way to ingest data from various sources and reducing the time and effort needed for integrations.
Grandata
Software Engineer Lead

As Lead Developer at Grandata, I managed the development of a personal data monetization platform, overseeing the team to ensure high-quality product delivery.

  • Managed a team of 8 people and was the main driver of technical decisions.
  • Promoted agile methodologies and team management practices, supporting effective product development.
  • Leadership, management, Scrum/agile, interviewing; React, React Native, NodeJS, AWS, GraphQL, MongoDB
Google
Customer Solutions Engineer

At Google, I worked as a Customer Solutions Engineer, providing technical support to sales teams by developing solutions for advertisers using the AdWords platform.

  • Collaborated across teams, using coding and communication skills to create prototypes, demos, and educational materials that generated significant revenue for Latin America's biggest companies (MercadoLibre, Despegar) and the Argentine Government.
  • Cross-team collaboration, customer facing, project management, digital advertisement; Python, Javascript, Angular, Google Cloud, AdWords
Grandata
Director of Data Infrastructure

As Director of Data Infrastructure at Grandata, I led the team responsible for providing real-time insights on large volumes of data for human analytics.

  • Main developer of an in-house analytics database implemented in C++, highly efficient — processing millions of rows per second on low-cost hardware.
  • Oversaw processing of data for hundreds of millions of users, focusing on optimizations to improve cost-effectiveness.
  • C++, LLVM, Python, Scala, Spark, Hadoop
Google
Systems Engineer Intern

During my internship at Google, I worked on the backend for reCaptcha, integrating StreetView data to improve image analysis at scale.

  • Developed a new approach to incorporate StreetView information into reCaptcha, speeding up the analysis of street images significantly.
  • C++, Google stack
GForge Group LLC
Systems Engineer

As the main developer at GForge, I redesigned the web-based collaboration tool used by companies like GE and Raytheon, focusing on architecture and optimization.

  • Designed the architecture for the new version and developed most of the core code, with extensions in Java and C#, leading to a stable system still in use today.
  • PHP, PostgreSQL, SCM tools (Git, SVN), Linux administration
Avatar S.A.
Intranet Systems Programmer

At Avatar, I designed and implemented corporate intranets for major Argentine companies, including TGS, the country's largest gas supplier.

  • Single-handedly implemented the e-commerce website for Fravega, Argentina's largest electronics retailer.
  • Addressed performance issues in the PHP framework by implementing effective caching solutions.

Skills

A broad toolkit spanning low-level systems to machine learning and team leadership.

Programming Languages
Ruby, Rust, C, C++, Python, Scala, PHP, SQL (MySQL, PostgreSQL, Yugabyte), x86 assembly, HTML, JavaScript, XML, Java, C#, Perl
Technologies & Frameworks
AI/LLMs, Machine Learning, Optimization, Distributed Systems, Big Data (Hadoop/HDFS, Spark, Cassandra, HBase), Libraries (Boost, STL, wxWidgets, QT, OpenCV, OpenGL, CUDA, OpenCL), Unix/Linux, SCM (SVN, Git, Mercurial), lex/yacc, MATLAB, Valgrind
Management & Methodologies
Team leadership, Scrum/Agile development, Cross-team collaboration, Project management

Honors & Memberships

PICT Scholarship (Scientific & Technology Research Program) IEEE Membership Argentine Astronomy Association Membership

Projects

Side projects exploring concurrency, databases, and productivity.

Elixir Concurrency
2023 – Present

LeechProxy

A proxy implemented in Elixir to filter unwanted websites during certain hours for improved productivity.

  • Built using Elixir for concurrency and fault-tolerance.
  • Enhances focus by blocking distractions during work hours.
C++ LLVM SQLite
2013 – 2016

Snel

Snel (SQL Native Execution for LLVM) is a database engine written from the ground up which uses SQLite as a frontend but compiles SQL queries into native code using the LLVM compiler frontend — providing blazingly fast SQL queries on low-cost hardware.

  • Built in C++ using the LLVM library.
  • Industry-proven.

Selected Publications

Peer-reviewed research in computer vision and pattern recognition.

Supervised Biometric System Using Multimodal Compression Scheme
W. Chaabane, R. Fournier, A. Naït-ali, J. Jacobo, M. Mejail, M. Mottalli, H. Ramos, A. Frery, P. Viana
17th Iberoamerican Congress on Pattern Recognition (CIARP 2012)
Read paper →
A Contrario Detection of False Matches in Iris Recognition
M. Mottalli, M. Tepper, M. Mejail
15th Iberoamerican Congress on Pattern Recognition (CIARP 2010)
Read paper →
Flexible Image Segmentation and Quality Assessment for Real-Time Iris Recognition
M. Mottalli, M. Mejail, J. Jacobo
Proceedings of the 2009 16th IEEE International Conference on Image Processing (ICIP), pp. 1941–1944
Read paper →
Real-time Identification of Persons using Iris Recognition
M. Mottalli
Master's Thesis (in Spanish)
Read thesis →

Education

Computer Science at the University of Buenos Aires.

University of Buenos Aires
Master's in Computer Sciences

Graduated with honors, achieving a stellar GPA of 8.62/10, showcasing academic excellence.

  • Thesis: Real-time Identification of Persons using Iris Recognition
  • Advisor: Dr. Marta E. Mejail
University of Buenos Aires
Bachelor's in Computer Sciences

Built a strong foundation in computer science, preparing for a career of innovation.

Interests

Cooking Photography Astronomy Running