Samantha Guerriero, Developer in London, United Kingdom
Samantha is available for hire
Hire Samantha

Samantha Guerriero

Verified Expert  in Engineering

Machine Learning Engineer and Developer

London, United Kingdom
Toptal Member Since
April 12, 2021

With over 20 data science and AI projects delivered for startups, enterprises, and academia, Sam offers a unique expertise blend as a senior AI consultant skilled in both hands-on engineering work and leadership roles with a special focus on team leading, project management, and thought leadership. By exploring the AI field from diverse perspectives and through varied projects, Sam is interested and committed to developing AI initiatives where growth, impact, and kindness are all at the center.


Artificial Intelligence (AI), AIOps, Machine Learning Operations (MLOps)...
Amazon Web Services (AWS), Snowflake, Data Build Tool (dbt), Dask, Python...
Google Cloud Platform (GCP), TensorFlow, Consulting, Presales, Recruiting...




Preferred Environment

Linux, MacOS, Google Cloud Platform (GCP), XGBoost, Scikit-learn, Keras, Python, Docker, Amazon Web Services (AWS), Kubernetes

The most amazing...

...project I've led is a CV model that identifies products from in-app images: >95% accuracy, >60% package-free product, improved UX, and lots of plastic saved!

Work Experience

AI Consultant

2021 - PRESENT
  • Delivered AI projects within specified timelines and budgets with measurable impact on increased efficiency, customer acquisition, and cost savings.
  • Maintained successful long-term relationships with a client retention rate of 30%.
  • Participated in webinars and meet-ups as an AI SME consulting for startups and small businesses.
  • Experienced the AI field from varied angles by contributing to technical, marketing, and educational projects as an individual contributor, project manager, advisor, and thought leader.
Technologies: Artificial Intelligence (AI), AIOps, Machine Learning Operations (MLOps), Machine Learning, Data Science, Google Cloud, Amazon Web Services (AWS), Thought Leadership, Advisory

Senior Machine Learning (MLOps) Engineer

2021 - 2022
  • Ideated and developed the automated retraining pipeline for the fraud engine model in Airflow.
  • Designed and set up an internal package for machine learning, MLOps, and data functionalities.
  • Introduced a new set of best practices for data, code, and model testing, leading to increased coverage by 50%.
  • Consulted periodically with the data science team on optimization for the current model pipeline.
  • Refactored multiple components of the deployment pipeline—from the feature engineering module to API service—which heavily reduced maintenance and running costs.
Technologies: Amazon Web Services (AWS), Snowflake, Data Build Tool (dbt), Dask, Python, TeamCity, Jenkins, Octopus Deploy, Linux, MacOS, XGBoost, Scikit-learn, Keras, Docker, Machine Learning, Statistical Methods, Database Design, Recruiting, Data Science, Machine Learning Automation, System Design, Cloud, Networking, Kubernetes, SQL, English, Automation, Google Structured Data, Batch Prediction, APIs, Deep Learning, Machine Learning Operations (MLOps), Content Writing, Artificial Intelligence (AI), Pandas, Big Data, Algorithms, Deep Neural Networks, Mathematics

Senior Machine Learning Engineer | Acting Head of AI Automation

2018 - 2021
  • Consulted, designed, and delivered DS, ML, and MLOps projects from PoC to production-ready systems with 100 NPS for clients like Realeyes, BT, Vodafone, and Lush.
  • Managed junior team members to set and achieve OKRs, fostering a proactive attitude towards achieving what they enjoy and want to learn.
  • Established the AI automation practice with the head of ML, from offerings and client profiles to strategic and marketing collateral for the internal team, clients, and stakeholders.
  • Cultivated and expanded the technical partnership with Intel through multiple R&D projects, resulting in research pieces with thousands of views online.
  • Initiated, coordinated, and participated in internal initiatives for the team, including all-hands days, Niko-Niko calendar, hiring, surveys, and OKR sessions.
  • Provided AI thought leadership pieces, white papers, talks, and training to practitioners and global leaders on best practices for AI on Google Cloud.
Technologies: Google Cloud Platform (GCP), TensorFlow, Consulting, Presales, Recruiting, Management, Thought Leadership, Computer Vision, Propensity Modeling, Deep Learning, Linux, MacOS, XGBoost, Scikit-learn, Keras, Kubeflow, Python, Docker, Machine Learning, Natural Language Processing (NLP), Statistical Methods, Business Planning, Database Design, Web Scraping, Sentiment Analysis, Data Science, Machine Learning Automation, Google Cloud Machine Learning, System Design, Cloud, Networking, Kubernetes, Cloud Dataflow, Pub/Sub, Google Kubernetes Engine (GKE), Cloud Storage, BigQuery, SQL, English, Automation, Google Structured Data, Batch Prediction, Google Cloud Video API, Google Cloud API, APIs, Object Detection, Data Build Tool (dbt), Dask, Google Cloud, Machine Learning Operations (MLOps), Content Writing, Artificial Intelligence (AI), Recommendation Systems, Pandas, Big Data, Algorithms, Deep Neural Networks, Mathematics, OCR, Blogging, Convolutional Neural Networks (CNN), Deep & Cross Network, Embedding Models

Machine Learning Researcher

2017 - 2017
University of Amsterdam
  • Designed the first deep version of NCM in literature, previously considered an impossible task by the community, and further enhanced the classifier with incremental learning and open-set learning techniques.
  • Obtained state-of-the-art accuracy for the NCM and Deep NCM algorithms in multiple scenarios.
  • Performed extensive analysis of state-of-the-art solutions and literature, further progressed by discussions in reading groups beside Ph.D. students.
  • Attended weekly meetings to present deliverables and propose new directions of work.
  • Gathered extensive experience in creating custom models in TensorFlow using graph mode and in Theano, following best practices in the design and finetuning of experiments.
Technologies: Research, Computer Vision, Python, TensorFlow, Deep Learning, Linux, Keras, Machine Learning, Statistical Methods, Mathematics, Web Scraping, Data Science, English, Batch Prediction, Content Writing, Artificial Intelligence (AI), Algorithms, Deep Neural Networks, Convolutional Neural Networks (CNN)

Machine Learning Intern

2015 - 2015
  • Researched and built a tailored method for automatic polarity tagging of sentences developed in RapidMiner.
  • Guided Wonderflow with the continuance of revenue generating by proposing tailor-built solutions and promoting the business’ technical advancement to investors alongside the founders of the company.
  • Researched and analysed competitors’ techniques and proposed improvements.
  • Reduced the business’ costs by 0.7€/review and optimized the business’ performance by 30 seconds/review, by conceptualizing and developing a new automatic mechanism for the business workflow.
  • Attended events alongside the founders and built relationships with other potential new business partners.
Technologies: Web Scraping, RapidMiner, Natural Language Processing (NLP), Sentiment Analysis, Machine Learning, Statistical Methods, Research, Data Science, English, Batch Prediction, Artificial Intelligence (AI), Algorithms, Mathematics

ML Pipeline Automation with Kubeflow

The client is a global company using Computer Vision to read the emotional responses of people looking at adverts online to understand which marketing campaigns have more impact. They were looking to scale up the efforts of their internal machine learning team to continuously improve their models and solutions by improving the way they create code, run experiments, and track/share results and serve their models.

Marketing Analytics - Brand Propensity Scoring

One of the UK’s largest retailers wanted to predict which of their thousands of brands customers are most likely to buy next. I developed a brand propensity model using TensorFlow and a variety of data sources, and A/B tested through an email campaign promoting new arrivals.

Object Detection for News Articles Retrieval

A fast-growing marketing and advertising agency looking to build a platform for clients and internal teams about where to search for news articles from multiple top newspaper websites. Starting from screenshots of the newspaper's main page, TF Object Detection is used to output stories (bounding boxes) within the page so to localize each news, and Google Cloud APIs are used to extract the origin website, image attached to the article header, text, keywords, and sentiment.

Technical Writer in Data and AI

• Published more than 40 articles and thought-leadership pieces as a ghostwriter for various data science, data, and AI companies.
• Improved the online engagement on the website for client companies by 3x.
• Researched topics ranging from TensorFlow Serving to ML model governance and data fabric.

AI Advisor for DS Platform Migration

• Developed a system requirement design document outlining the business use case, architecture design, and functional requirements of lift and shifting the current infrastructure to Kubeflow.
• Consulted on the implementation plan for the lift and shift of the infrastructure, with a PoC delivered by the team in two weeks.
• Advised on the pros and cons of different solutions to Kubeflow, focusing on Vertex AI and other cloud solutions.

AI Consultant for Startup

• Conceptualized the matching solution for the launch of an online platform connecting talent to job opportunities.
• PM-ed and contributed to the creation of an initial rule-based solution that could be used from day one and the definition of new UX elements for data gathering.
• Created a roadmap detailing how and when to move to a deep-learning solution for the matching engine.

Recommender System for a Large Online Retailer

Developed a model for RecSys built on millions of user interactions. Experimented with matrix factorization (as baseline) and different deep learning approaches for collaborative filtering, including wide, deep, and tower models, trained on a tuned feature set, including hand-engineered feature crosses. The final model training process was delivered as an automated pipeline from data processing to model evaluation.
2015 - 2017

Master's Degree in AI & Robotics

Sapienza - Rome, Italy

2012 - 2015

Bachelor's Degree in Information Technology

RomaTre - Rome, Italy


Machine Learning Engineering for Production (MLOps)

DeepLearning.AI | via Coursera


Google Cloud Certified Professional Machine Learning Engineer

Google Cloud


Google Cloud Certified Cloud Architect

Google Cloud


Google Cloud Certified Data Engineer

Google Cloud


Cambridge English: Proficiency (CPE)

University of Cambridge


TensorFlow, Pandas, XGBoost, Scikit-learn, Keras, Google Cloud Video API, Google Cloud API, Dask, PyTorch


Jupyter, Amazon SageMaker, 2Checkout, Cloud Dataflow, Google Kubernetes Engine (GKE), BigQuery, TeamCity, Jenkins, Google AI Platform, Azure Machine Learning, Jira


Python, SQL, Snowflake, JavaScript


Data Science, Database Design, Management, Automation, Agile Software Development, DevOps, Agile


Google Cloud Platform (GCP), Kubeflow, Linux, Docker, MacOS, RapidMiner, Kubernetes, Amazon Web Services (AWS)


Google Cloud




Machine Learning, Computer Vision, Consulting, Thought Leadership, Artificial Intelligence (AI), Recommendation Systems, Algorithms, Deep Neural Networks, Neural Networks, Presales, Sentiment Analysis, Machine Learning Automation, Google Cloud Machine Learning, Object Detection, Deep Learning, Content Writing, Image Processing, Big Data, Technical Writing, Convolutional Neural Networks (CNN), OCR, Image Recognition, Writing & Editing, Blogging, AI Democratization, Natural Language Processing (NLP), Robotics, Statistical Methods, Mathematics, Business Planning, Economics, Recruiting, Propensity Modeling, Research, Web Scraping, System Design, Cloud, Networking, Pub/Sub, Cloud Storage, English, Google Structured Data, Batch Prediction, APIs, Data Build Tool (dbt), Octopus Deploy, Machine Learning Operations (MLOps), Technical Design, Large Language Models (LLMs), Diffusion Models, Generative Adversarial Networks (GANs), Data Collection, Technical Requirements, User Experience Design, Product Roadmaps, AIOps, Advisory, Retrieval-augmented Generation (RAG), Deep & Cross Network, Embedding Models, Matrix Factorization, Feature Engineering, Collaborative Filtering, Model Tuning

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.


Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring