Andrew C.

AI & Machine Learning Engineer • Paris, IDF, FR

Natural Language Processing

Statistics

Data Science

Machine Learning

Deep Learning

PyTorch

SQL

Python

Analytics

Software Development

Mathematics

Computer Science

Optimization

Big Data

Data Analysis

Forecasting

Cloud Computing

Data Visualization

Algorithms

Data Mining

NumPy

Statistical Analysis

Amazon Web Services

Linux

Google Cloud Platform

Docker

Databases

MySQL

Git

Tensorflow

Communication

English

Computer Vision

Programming

Project Management

C++

Available for hire fromNegotiable

Contracts

Full-Time Roles

Data Scientist & Machine Learning Engineer | NLP, Statistical Analysis & Deep Learning | SQL & Python | Predictive Modeling, Data Mining & Analytics

- I have over 10 years of experience in Data Science & Analytics, with 7 years dedicated to Python & Machine Learning and 5 years to Deep Learning & Engineering. - My recent focus has been on NLP, where I have honed my skills in BERT, Transformers, Transfer Learning, and LLM APIs. - I have a strong foundation in machine learning methodologies and statistics. - My expertise includes SQL, managing large datasets, API development, and deploying solutions on GCP/AWS. - I am proficient in communicating complex data insights and enjoy mentoring others.

Projects

AutoML: Unsupervised Model Training with Optuna & SHAP Feature Selection

- Objective: Aimed to build an AutoML platform, leveraging seasoned expertise and the best methodologies for optimal performance. - Approach: Focused on refining hyperparameter optimization for greater efficiency, innovating in feature selection for more profound results, and ensuring development integrity through rigorous testing, making the system robust and user-friendly. - Result: The platform now effectively supports essential model types like linear, gradient boosting, and Naive Bayes, with ongoing enhancements to broaden its modeling capabilities and application scope.See More

Web Scraping: Efficient Data Collection Using Selenium and Headless Chrome

- Objective: To create a web scraping tool capable of gathering league ranking data from TeamForm, utilizing a headless Chrome browser for efficient data collection. - Approach: Deployed Selenium for automated web navigation and scraping, with added functionality to manage data load for memory efficiency. The design also allows for future expansion to collect more detailed 'Club' and 'National' data. - Result: Successfully extracted league ranking data, offering valuable insights into team standings and performance, and set the groundwork for expanded data retrieval capabilities.See More

Algorithm: DTW-Based Hierarchical Clustering for FMCG Sales Time Series Analysis

- Objective: To unlock insights within Consumer Goods sales data through detailed analysis and clustering to highlight patterns and trends. - Approach: Conducted thorough time series analysis, including data cleaning for quality and employing Dynamic Time Warping (DTW) to pinpoint similarities, alongside developing a NumPy-based clustering algorithm for efficient data aggregation. - Result: Successfully clustered over 10,000 time series data points, revealing meaningful patterns and trends, significantly enhancing data understanding and strategic planning capabilities.See More

Work History

Data Scientist & ML Engineer: Strategic Data Insights & Data Enrichment | Data-Driven Analytics

CREDNov 2020 - Oct 2023 • 3 yrsBrought over 25 projects from ideation to deployment, including several foundation models, delivering data-driven insights and diversifying the company’s product lineup. - Developed multimodal enrichment and standardization models using LLMs, restoring up to 90% of missing data (~450M data points); these initiatives expanded the volume and accuracy of data deliverables, directly impacting revenue growth by enhancing data quality for client projects. - Collaborated on 10+ sports analytics models that professional scouts and clubs adopted; these models optimized scouting processes and costs and featured in mobile apps and dashboards. - Initiated data quality initiatives, including a game statistics pipeline that improved the precision of key predictive models, impacting the accuracy of player performance assessments. - Built statistical models that introduced matching/recommendations between 100,000+ businesses and 400M consumers, directly reducing customer attraction costs. - Established a dual-layer regression model for football market value forecasts with <10% error, improving strategic decision-making for industry experts. - Deployed an athlete retirement prediction model with a one-season margin of error, managing data irregularities and guiding strategic investments. - Engineered a salary prediction pipeline with less than 10% error, providing insights that deepened client understanding of consumer demographics. - Innovated algorithms for customer data refinement, enhancing segmentation and impacting ad campaigns, reducing marketing expenditures. - Facilitated development of a Streamlit dashboard with complex visuals and analytics, optimizing client presentations. - Leveraged Generative AI to analyze and process 10M social media profiles, adding depth to our datasets.

DS & ML Consultant: Predictive & Time Series Analysis | Algorithm Design & NLP

Independent Consulting ServicesJan 2017 - Jul 2022 • 5 yrs 7 mosCollaborated with emerging businesses (startups & small companies) and consulting companies across various sectors to develop ML solutions from the ground up, spanning predictive analytics, time series forecasting, algorithm creation, and NLP. These solutions demonstrated direct and indirect cost savings efficiencies and influenced revenue growth. In addition, this role demanded extensive teamwork, self-project management, and a deep understanding of each client's needs. - Created an Employee Churn Detection Model through close collaboration with the HR department, offering targeted and actionable insights; this initiative, highly praised by stakeholders, boosted retention rates and identified more than 10 important at-risk employees over the project's duration. - Developed a PDF processing algorithm that extracts structured data from diverse financial documents dating back 10 years (totalling over 1k papers); this achieved an accuracy rate of over 95%, enabling custom search functionalities. - Innovated a Speech Anomaly Detection Algorithm with 70-95% accuracy across more than 40 defect types; this solution, resulting from collaborative R&D efforts, implemented a core update to a healthcare mobile app. - Devised a Horse Race Betting System offering real-time, low-latency betting suggestions; this method doubled the performance of the previous system and resulted in an approximate 2% increase in revenue. - Created a sales volume clustering algorithm, which led to a 15% improvement in sales planning and effectiveness.

Middle/Senior Data Analyst: Analytics & Strategy | Modeling Impact & Revenue Growth

Association 'Non-Profit Market Council'Feb 2012 - Dec 2016 • 4 yrs 11 mosPlayed a pivotal role in crafting and implementing data-driven strategies, managing data processing and analytical modeling. As a Senior Analyst, led projects that boosted clients' decision-making capabilities, yielding revenue increases between 2.5% and 10% in the following year. - Mentored and led 3 junior analysts, fostering a learning and professional growth culture; the guidance facilitated skill enhancement and resulted in two promotions within the year, demonstrating commitment to team development and leadership. - Revolutionized operational efficiency within the team, reducing task completion time from 4 days to 8 hours through improving automation scripts, impacting the department's proficiency in delivering quick and accurate reports and insights. - Applied time series analysis and data science techniques for anomaly detection, identifying around 50 critical periods annually; this led to a 5-25% reduction in forecast error rates for power price/volume predictions in targeted regions. - Introduced a data enrichment algorithm, aggregating daily data into weekly and monthly summaries; this innovation improved analysis accuracy during volatile periods, contributing to more reliable forecasting models. - Promoted to Senior Analyst in 2014 for exceptional predictive modeling expertise and productivity enhancements, having developed over 10 models that influenced strategic decisions and operational efficiency. - Pitched and received approval for implementing 5 predictive models in over 20 stakeholder meetings by communicating their technical and business impacts.

Junior Data Analyst: Data Processing & Analytical Modeling | ROI | SQL

BrandScienceJul 2011 - Feb 2012 • 8 mosFacilitated execution of data-centric strategies, focusing on data processing and analytical modeling and driving insights influencing strategic decisions. - Developed and implemented SQL and VBA-based aggregation logic, incorporating correlation analysis to expand data by 2.5x and enable multiple data sources for modeling, thereby enhancing reliability and supporting more informed decision-making. - Streamlined media data collection (CATI/CAWI) and processing by introducing automated scripts, increasing departmental task efficiency by 300% and reducing algorithm execution time from 2 hours to 30 minutes while making it fully automatic. - Initiated cluster analysis to estimate early-stage campaign efficiency, enabling more strategic budget pre-allocation; this approach improved budget allocation effectiveness by an average of 20% across over 10 campaigns. - Enhanced ROI models by integrating a VBA-based anomaly detection function, stabilising predictions at the early forecasting stages, reducing expenditures by 50%, and boosting client marketing budget efficiency. - Collaborated in refining the ROI prediction regression model using Excel/VBA, boosting campaign efficiency by 10% (improving brand knowledge from approximately 80% to 90%).

Education

Peoples’ Friendship University of Russia

Master of Science - MS, Applied Mathematics & Computer ScienceSep 2004 - Jul 2011

Similar Pangeans

How Pangea Works

Effortlessly discover top talent

We’ve distilled the candidate search from endless hours down to just a few minutes. Using Pangea’s AI-powered search tools, you can find top fractional talent able to take on your next project. Our system looks at your company’s niche and your needs to find the perfect match faster than any traditional hiring platform.

Start working with talent today

The top talent on Pangea is ready to get started with you right now. You can message or hire a candidate right from their profile page and start assigning work as soon as they respond. And the best part? Pangea’s fractional contract structure lets you start small and ramp up as your needs change, keeping your costs manageable and your team’s capabliities flexible.

Track work and invoices in one place

Assign tasks, track progress, and complete invoices all on Pangea. We’ve combined every part of the hiring process into one platform to eliminate the miscommunication that’s unavoidable on other freelance platforms. We even send out 1099s to your contractors at the end of the year!

Talk with a Talent Expert

Members of our team are available to help you speed through the hiring process.

Available Now

Book a Call