Claudio is available for hire
Hire ClaudioClaudio S. De Mutiis
Verified Expert in Engineering
Data Science and Machine Learning Developer
London, United Kingdom
Toptal member since January 21, 2021
Bio
Claudio is a senior data scientist with experience in stakeholder management, recruitment, and line management. He is proficient in supervised, unsupervised, and reinforcement learning, including deep learning and neural networks. He has worked on several applications of machine learning and AI, including NLP and computer vision. Claudio has extensive industry experience, including retail/eCommerce, media, high-tech, startups, insurance, and healthcare.
Portfolio
MINORO LTD.
Python 3, Python, Scikit-learn, Google Cloud Platform (GCP), BigQuery...
Bilt Technologies Inc
Machine Learning, Python, Supervised Learning...
Carrier Global
Python, Data Science, Predictive Modeling, Machine Learning...
Experience
- Computer Vision - 9 years
- Data Science - 9 years
- SQL - 8 years
- Python - 8 years
- Machine Learning - 8 years
- Natural Language Processing (NLP) - 7 years
- Keras - 5 years
- Scikit-learn - 5 years
Preferred Environment
Amazon Web Services (AWS), Python, MATLAB, Jupyter Notebook, Snowflake, Keras, TensorFlow, Scikit-learn, Pandas, SQL
The most amazing...
...thing I've built and trained is a convolutional neural network for end-to-end driving in a simulator using Keras.
Work Experience
Senior Data Scientist
2025 - 2025
MINORO LTD.
- Spearheaded a sales forecasting project for a large client in the finance space.
- Trained and evaluated overall sales forecasting models by region.
- Trained and evaluated sales forecasting models tailored to specific business partners.
Technologies: Python 3, Python, Scikit-learn, Google Cloud Platform (GCP), BigQuery, Google BigQuery, Pandas, GitHub, TensorFlow, Regression, random forest, Forecasting, Leadership, Datasets
Machine Learning Engineer
2024 - 2024
Bilt Technologies Inc
- Contributed to fine-tuning a GPT-4 model for customer benefit personalization.
- Built a vector store POC to be used as part of a retrieval-augmented generation (RAG) system using LlamaIndex and ChromaDB.
- Built a vector store in Google BigQuery, which allows querying vector embeddings and location metadata, including latitude and longitude, at the same time.
- Performed a merchant and benefit data analysis and built a dashboard with Evidence to identify weaknesses and opportunities for improvements.
Technologies: Machine Learning, Python, Supervised Learning, Natural Language Processing (NLP), Large Language Models (LLMs), LangChain, LlamaIndex, ChromaDB, OpenAI API, OpenAI GPT-4 API, LLM Fine-tuning, Streamlit, Evidence, Google BigQuery, Vector Stores, Retrieval-augmented Generation (RAG), Data Cleansing, OpenAI, Datasets
Data Science Lead
2024 - 2024
Carrier Global
- Managed a team of three data scientists and one data engineer. Greatly improved the data science team's workflow, internal communication, and knowledge sharing.
- Spearheaded a very large data migration from Amazon S3 files and other data sources to Snowflake, which made data investigations more efficient, increased data transparency, and empowered data science and analytics initiatives.
- Oversaw the delivery of a POC regarding detecting energy outliers based on other users with similar characteristics. The data output was dropped in production with a regular service so the consumer app team could pick up and use it.
- Led the delivery of a POC to forecast user energy consumption based on various features, including weather forecast, temperature setpoints, and hardware characteristics.
Technologies: Python, Data Science, Predictive Modeling, Machine Learning, Amazon Web Services (AWS), Kubernetes, Energy Management, Kubeflow, Prefect, Snowflake, Data Migration, K-nearest Neighbors (KNN), Docker, Jupyter Notebook, Supervised Learning, Supervised Machine Learning, Data Analysis, Exploratory Data Analysis, Data Cleansing, Leadership, Datasets
Data Scientist
2023 - 2023
Sema Technologies, Inc
- Brainstormed with another data scientist and wrote documentation on how to discriminate between code written by AI and AI blended code, i.e. partly human and partly AI.
- Wrote Python parsers and scripts to gather and process source code datasets from GitHub in a format appropriate to be fed to our model. Those datasets were split into human and generative AI codes for binary classification.
- Researched and pointed out various sources of model overfitting within the data science team and to my manager.
Technologies: Machine Learning, Data Science, Large Language Models (LLMs), Natural Language Processing (NLP), Generative Artificial Intelligence (GenAI), Data Cleansing, OpenAI, Datasets
Lead Data Scientist
2023 - 2023
Wilmington plc
- Researched and developed forecasting models for the gross written premiums of several insurance lines across the globe (Axco Insurance).
- Researched and developed predictive models for patient numbers for different therapy areas and drugs across the UK (Wilmington Healthcare).
- Contributed to a proof of concept involving the use of large language models (LLMs) for grading and student feedback (Wilmington Training and Education).
- Collaborated with the innovation committee at Wilmington.
Technologies: Python 3, Scikit-learn, Pandas, Jupyter Notebook, Jupyter, Snowflake, Tableau, Citrix, SharePoint, Jira, Confluence, Data Science, Forecasting, Predictive Modeling, Machine Learning, Artificial Intelligence (AI), ChatGPT, Data Visualization, Frameworks, Generative Pre-trained Transformers (GPT), Data Scientist, Regression Modeling, Quantitative Analysis, OpenAI GPT-3 API, OpenAI API, Language Models, Large Language Models (LLMs), Data Wrangling, ARIMAX, Statistical Data Analysis, Generative Artificial Intelligence (GenAI), Data Cleansing, Leadership, OpenAI, Datasets
Lead Data Scientist
2022 - 2023
Winnow
- Managed and mentored a team of data scientists. Led strategic projects/insights.
- Supported the development of computer vision models.
- Improved and automated the workflow and processes.
- Improved coding practice and reviewed existing projects' code.
- Managed the Agile way of working and sprint planning within the data science team. Also managed annotation and data quality.
Technologies: Python 3, Python, SQL, Jupyter Notebook, Mode Analytics, Data Science, Statistics, Pandas, Jira, Slack, Agile, Sprint Planning, Data Visualization, Data Scientist, Quantitative Analysis, Data Wrangling, Statistical Data Analysis, Computer Vision, Computer Vision Algorithms, Data Cleansing, Leadership, Datasets
Senior Data Scientist
2021 - 2022
Sky UK
- Developed and implemented customer churn models for the business.
- Collaborated on a real-time machine learning proof of concept involving anomaly detection on hub telemetry data.
- Interviewed and recruited data scientist candidates and fulfilled line management responsibilities.
Technologies: Churn Analysis, Real-time Data, Anomaly Detection, Google Cloud Platform (GCP), BigQuery, Jupyter Notebook, Python 3, Keras, TensorFlow, Scikit-learn, Pandas, Supervised Learning, Supervised Machine Learning, Unsupervised Learning, Autoencoders, Google Cloud Storage, Google BigQuery, Data Science, Machine Learning, Python, SQL, Classification, Clustering, Information Retrieval, Algorithms, Artificial Intelligence (AI), Neural Networks, Deep Learning, Data Analysis, Data Analytics, Data Reporting, Big Data, Data, GitHub, Data Preprocessing, Document Processing, Feature Engineering, Data Processing, Deep Neural Networks (DNNs), Jira, Slack, Agile, Sprint Planning, Data Visualization, Frameworks, Data Scientist, Quantitative Analysis, Data Wrangling, Statistical Data Analysis, Data Cleansing, Leadership, Datasets
Senior Data Scientist
2021 - 2021
Integral Solutions, Inc.
- Investigated the pros and cons of using different NBA APIs.
- Wrote scripts to retrieve NBA data, process it, and store it on S3.
- Performed feature engineering using team stats and other handcrafted features coming from historical NBA matches.
- Designed, validated, and tested a deep neural network model to predict NBA winners and losers as well as the winning probabilities.
- Collaborated with a software engineer to put the NBA prediction model in production for the first MVP of the project.
- Achieved market-leading accuracy for predicting NBA match outcomes.
- Wrote some documentation, introduced some unit tests, and suggested future developments for the project and actions that could further improve the existing model.
Technologies: Python, Amazon S3 (AWS S3), Amazon Web Services (AWS), Deep Learning, Deep Neural Networks (DNNs), Neural Networks, Predictive Modeling, Data Processing, APIs, Minimum Viable Product (MVP), Documentation, Feature Engineering, Churn Analysis, Data Science, Machine Learning, Scikit-learn, Keras, Pandas, Classification, Jupyter Notebook, Algorithms, Artificial Intelligence (AI), Supervised Learning, Data Analysis, Data Analytics, Data Reporting, Data, GitHub, Data Preprocessing, Document Processing, Frameworks, Data Scientist, Quantitative Analysis, Data Wrangling, Statistical Data Analysis, Data Cleansing, Datasets
Senior Data Scientist
2018 - 2019
Notonthehighstreet Enterprises Ltd
- Managed a topic classification NLP project using convolutional neural networks and word embeddings to be used by the partners and operations/customer service team.
- Led a deep neural network recommender system project that led to valuable customer segmentation insights to be used by the product and curation team.
- Collaborated with the digital marketing team to increase the effectiveness of marketing and advertising campaigns as well as SEO.
- Managed a competitor analysis project to be used as insights by the executive team.
- Improved data science workflow and coding practices.
- Redesigned the data science recruitment from scratch.
- Managed, guided, and mentored a mid-level data scientist.
- Built a product bundles graph to visualize insights on products frequently bought together.
- Documented data science projects on a data team wiki.
- Managed a multi-touch digital marketing attribution project using a Markov chain.
Technologies: Big Data, Data, R, Dimensionality Reduction, Unsupervised Learning, Supervised Learning, Neural Networks, eCommerce, Data Analytics, Document Processing, Google Analytics, Google SEO, Algorithms, Recommendation Systems, Continuous Integration (CI), Data Reporting, Data Analysis, Conversion Rate, Machine Learning, Data Science, Amazon Web Services (AWS), GitHub, Statistical Analysis, Mesos, Relational Databases, Tableau, Pattern Recognition, Calculus, Linear Algebra, Word Embedding, Convolutional Neural Networks (CNNs), Snowflake, Clustering, Information Retrieval, Classification, Regression, Natural Language Processing (NLP), Artificial Intelligence (AI), Pandas, Python, Jira, Slack, LaTeX, CHRONOS, Jenkins, Ansible, Jupyter Notebook, Scikit-learn, Keras, TensorFlow, SQL, Python 3, Writing & Editing, Documentation, Jupyter, Data Modeling, Statistics, Matplotlib, NumPy, Deep Learning, Chatbots, APIs, Git, Docker, Text Analytics, Data Preprocessing, Amazon EC2, MySQL, Classification Algorithms, Amazon S3 (AWS S3), Churn Analysis, Feature Engineering, Data Processing, Deep Neural Networks (DNNs), Agile, Sprint Planning, Data Visualization, Frameworks, Data Scientist, Regression Modeling, Quantitative Analysis, Search Engines, Data Wrangling, Retail & Wholesale, Statistical Data Analysis, Marketing, Data Cleansing, Leadership, Datasets
Data Scientist
2017 - 2018
Notonthehighstreet Enterprises Ltd
- Worked on an NLP semantic search project using word embeddings in collaboration with tech and other product stakeholders.
- Built predictive models to evaluate our business partners' success to be used as actionable insights by the partners and operations team.
- Engaged and built relationships with senior stakeholders throughout the business.
- Worked on an external trending/social media influencers/posts ranking project in collaboration with the product and curation team that led to the development of a web app to make their job easier.
- Contributed to creating and introducing a data team learning and development culture.
- Placed an NLP project in production to detect a set of specific things in messages business partners sent to customers to be used as actionable insights by the partners and operations team and to be included in a weekly report.
- Documented data science projects on a data team wiki.
Technologies: Big Data, Data, R, Dimensionality Reduction, Unsupervised Learning, Supervised Learning, Neural Networks, eCommerce, Data Analytics, Document Processing, Algorithms, Continuous Integration (CI), Data Reporting, Data Analysis, Conversion Rate, Machine Learning, Data Science, Amazon Web Services (AWS), GitHub, Statistical Analysis, Jira, Slack, Relational Databases, Pattern Recognition, Calculus, Linear Algebra, Word Embedding, Convolutional Neural Networks (CNNs), Clustering, Information Retrieval, Classification, Regression, Natural Language Processing (NLP), Artificial Intelligence (AI), Pandas, Python, LaTeX, Mesos, CHRONOS, Jenkins, Ansible, Jupyter Notebook, Scikit-learn, Keras, TensorFlow, SQL, Python 3, Writing & Editing, Documentation, Jupyter, Data Modeling, Statistics, Matplotlib, NumPy, Deep Learning, APIs, Git, Docker, Text Analytics, Data Preprocessing, Amazon EC2, MySQL, Classification Algorithms, Amazon S3 (AWS S3), Feature Engineering, Data Processing, Deep Neural Networks (DNNs), Data Visualization, Frameworks, Data Scientist, Regression Modeling, Quantitative Analysis, Search Engines, Data Wrangling, Retail & Wholesale, Statistical Data Analysis, Data Cleansing, Datasets
Data Scientist
2016 - 2016
Mindi Technologies Ltd
- Wrote Python scripts to analyze 36 features of DigitalOcean's servers' data such as droplets_cpu_stime, droplets_cpu_utime, droplets_network_rxbytes, and droplets_network_txbytes.
- Worked with the server's droplets of nine different sizes (512 MB, 1GB, 2GB, 4GB, 8GB, 16GB, 32GB, 48GB, and 64GB).
- Tried to infer server and droplet power usages from the datasets provided by DigitalOcean.
Technologies: Big Data, Data, Data Analytics, Data Analysis, Statistics, Matplotlib, NumPy, Python 3, SQL, MySQL, Data Processing, Data Visualization, Data Scientist, Quantitative Analysis, Data Wrangling, Statistical Data Analysis, Data Cleansing, Datasets
AI Researcher
2016 - 2016
King's College London
- Worked on a project that was part of a collaboration between researchers in artificial intelligence, telecommunications, and environmental sciences. The project was carried out in partnership with Transport for London (TFL) and Ericsson.
- Used artificial intelligence planning to contribute to the design of the next generation of intelligent urban traffic controls (i.e., AI-controlled traffic lights, speed limits, and route planning).
- Visited the TFL operational center and learned about the SCOOT system. Learned about traffic systems used in other main cities around the world.
- Studied the papers written by some of the world's most prominent research groups on traffic optimization.
- Used traffic simulation tools such as SUMO (simulation of urban mobility) and PTV Vissim to simulate congestion scenarios in London.
- Wrote Python scripts that were part of the framework used to interface the DINO AI planner and SUMO.
- Attended the 26th International Conference on Automated Planning and Scheduling (ICAPS 2016) in London.
- Guided and mentored a couple of students in the Master of Science degree program.
Technologies: Artificial Intelligence (AI), Python, Planning, APIs, Git, Data Visualization, Quantitative Analysis, Data Wrangling, Datasets
Experience
Advanced Lane Finding
Built an advanced lane-finding algorithm using distortion correction, image rectification, color transforms, and gradient thresholding. Identified lane curvature and vehicle displacement. Overcame environmental challenges, such as shadows and pavement changes.
Vehicle Detection and Tracking
Created a vehicle detection and tracking pipeline with OpenCV, a histogram of oriented gradients (HOG), and support vector machines (SVM). Optimized and evaluated the model on video data from an automotive camera taken during highway driving.
Using Deep Learning to Clone Driving Behavior
Built and trained a convolutional neural network for end-to-end driving in a simulator, using Keras. Used optimization techniques such as regularization and dropout to generalize the network for driving on multiple tracks.
Traffic Sign Classification
Built and trained a deep neural network to classify traffic signs using TensorFlow. Experimented with different network architectures. Performed image pre-processing and validation to guard against overfitting.
Local Odometry Techniques for a Differential Wheeled Robot
• Provided the MIRTO robot, designed and built by a team led by Dr. Franco Raimondi at Middlesex University London, with autonomous navigation planning capabilities.
• Wrote a library of high-level odometrical functionalities (i.e., Java methods) to allow MIRTO to perform actions such as rotating, translating, and moving towards a specific point in space while avoiding all obstacles on the way.
• Developed a navigation algorithm that only uses MIRTO's wheels' encoders and bumpers sensors.
• Used MIRTO to test the newly developed navigation algorithm.
• Discussed the results and observations in my MSc project's dissertation.
• Wrote a library of high-level odometrical functionalities (i.e., Java methods) to allow MIRTO to perform actions such as rotating, translating, and moving towards a specific point in space while avoiding all obstacles on the way.
• Developed a navigation algorithm that only uses MIRTO's wheels' encoders and bumpers sensors.
• Used MIRTO to test the newly developed navigation algorithm.
• Discussed the results and observations in my MSc project's dissertation.
Predicting Boston Housing Prices
Built a model to predict a given house's value in the Boston real estate market using various statistical analysis tools. Identified the best price that a client can sell their house utilizing machine learning.
Education
2014 - 2015
Master of Science Degree in Robotics
King’s College London - London, United Kingdom
2009 - 2013
Bachelor of Arts Degree in Physics
Boston University - Boston, MA, USA
2009 - 2013
Bachelor of Arts Degree in Economics
Boston University - Boston, MA, USA
Certifications
OCTOBER 2023 - PRESENT
Natural Language Processing with Probabilistic Models
DeepLearning.AI | via Coursera
SEPTEMBER 2023 - PRESENT
Natural Language Processing with Classification and Vector Spaces
Coursera
SEPTEMBER 2023 - PRESENT
Practical Time Series Analysis
The State University of New York | vis Coursera
NOVEMBER 2018 - PRESENT
Machine Learning: Clustering & Retrieval
Coursera
NOVEMBER 2018 - PRESENT
Machine Learning
Stanford University via Coursera
NOVEMBER 2018 - PRESENT
Machine Learning Specialization
University of Washington via Coursera
DECEMBER 2017 - PRESENT
Machine Learning: Classification
Coursera
FEBRUARY 2016 - PRESENT
Machine Learning: Regression
Coursera
Skills
Libraries/APIs
Pandas, Scikit-learn, TensorFlow, Keras, OpenCV, Matplotlib, NumPy, OpenAI API, Camera API
Tools
MATLAB, LaTeX, Slack, Jira, GitHub, Jupyter, Git, BigQuery, ARIMAX, Google Analytics, Tableau, Ansible, Jenkins, Mesos, Confluence, ChatGPT, ARIMA, Prefect
Languages
SQL, Snowflake, Java, Python, Python 3, C++, C, Fortran, R
Paradigms
Real-time Systems, Anomaly Detection, Agile, MapReduce, Continuous Integration (CI)
Platforms
Jupyter Notebook, Amazon Web Services (AWS), Docker, Amazon EC2, Google Cloud Platform (GCP), Citrix, SharePoint, Kubernetes, Kubeflow
Storage
Relational Databases, MySQL, Amazon S3 (AWS S3), Google Cloud Storage
Industry Expertise
Banking & Finance, Retail & Wholesale, Marketing
Frameworks
LlamaIndex, Streamlit
Other
Linear Algebra, Calculus, Differential Equations, Computational Physics, Quantum Physics, EM Waves, Mechanics, Physics, Microeconomics, Macroeconomics, Game Theory, Labor, Complex Variables, Theoretical Physics, Computer Vision, Artificial Intelligence (AI), Multi-agent Systems, Biotechnology, Sensors & Actuators, Pattern Recognition, Regression, Classification, Information Retrieval, Expectation-Maximization (EM), Natural Language Processing (NLP), Word Embedding, Reinforcement Learning, Self-driving Cars, Color Grading, SVMs, Convolutional Neural Networks (CNNs), Robotics, Navigation, Visual Odometry, Data Science, Machine Learning, Conversion Rate, Data Analysis, Data Reporting, Algorithms, Recommendation Systems, Document Processing, Neural Networks, Unsupervised Learning, Dimensionality Reduction, Data Analytics, eCommerce, Supervised Learning, Data, Big Data, Writing & Editing, Documentation, Data Modeling, Statistics, Deep Learning, Image Recognition, Object Tracking, APIs, Data Preprocessing, Text Analytics, Classification Algorithms, Deep Neural Networks (DNNs), Predictive Modeling, Data Processing, Minimum Viable Product (MVP), Feature Engineering, Churn Analysis, Real-time Data, Supervised Machine Learning, Autoencoders, Google BigQuery, Sprint Planning, Monetary Policy, Clustering, Environment, Mathematics, Generative Pre-trained Transformers (GPT), Forecasting, Data Visualization, Frameworks, Data Scientist, Regression Modeling, Quantitative Analysis, OpenAI GPT-3 API, Language Models, Search Engines, Data Wrangling, Statistical Data Analysis, Computer Vision Algorithms, Data Cleansing, Leadership, OpenAI, Datasets, Calibration, Google SEO, Optical Character Recognition (OCR), Chatbots, Large Language Models (LLMs), Generative Artificial Intelligence (GenAI), CHRONOS, Statistical Analysis, Mode Analytics, Economics, Planning, Time Series, Time Series Analysis, Machine Translation, Locality-Sensitive Hashing, Sentiment Analysis, Vector Space Models, Linear Regression, Ridge Regression, Lasso Regression, Logistic Regression, Decision Trees, K-means Clustering, K-D Tree, Word2Vec, Parts-of-Speech Tagging, POS, N-gram Language Models, Continuous Bag of Words (CBOW), Autocomplete, Clustering Algorithms, Energy Management, Data Migration, K-nearest Neighbors (KNN), Exploratory Data Analysis, LangChain, ChromaDB, OpenAI GPT-4 API, LLM Fine-tuning, Evidence, Vector Stores, Retrieval-augmented Generation (RAG), random forest
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.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
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