Carleton Coggins
Verified Expert in Engineering
Data Scientist and Developer
Nashville, TN, United States
Toptal member since September 24, 2025
Carleton is a versatile data scientist with nearly a decade of experience delivering high-impact ML, AI, and analytics solutions. His expertise spans business intelligence through production AI systems, helping businesses innovate, grow, and make a positive impact. Carleton recently led the development of an enterprise predictive intelligence platform serving millions of users, driving significant revenue growth and market differentiation.
Portfolio
Experience
- Python - 10 years
- Natural Language Processing (NLP) - 7 years
- Business Intelligence (BI) - 7 years
- SQL - 7 years
- Deep Learning - 5 years
- Machine Learning - 5 years
- Large Language Models (LLMs) - 2 years
- Retrieval-augmented Generation (RAG) - 1 year
Preferred Environment
SQL, Python, Machine Learning, Business Intelligence (BI), Artificial Intelligence (AI)
The most amazing...
...project I've done involved building Crunchbase's flagship funding predictor, now used by millions to predict market patterns.
Work Experience
Senior Data Scientist
CrunchBase
- Led the development of advanced AI/ML features to Crunchbase’s private market predictive intelligence platform, contributing significant revenue growth and establishing market differentiation.
- Designed a flagship hybrid forecasting model to predict the likelihood of a funding event—achieving 95% precision and 99% recall—empowering investor and sales professionals to identify high-potential startups.
- Developed a high-precision workforce reduction model, providing strategic risk insights for market trend analysis.
- Pioneered the creation of reusable data assets, enabling long-term innovation and reducing development cycles by around 25%.
- Increased subscription trial starts across millions of users by developing a two-stage recommendation engine with approximate nearest neighbors (ANN) for candidate generation and PyTorch for ranking.
- Applied large-language models (LLMs) and prompt engineering to analyze investor portfolio patterns, enabling targeted startup–investor matching.
- Analyzed customer engagement across 5+ million companies to create a ranked "golden set" for targeted LLM feature deployment, establishing organization-wide standards for optimal budget allocation.
- Developed a churn forecasting model based on user engagement patterns, identifying at-risk customers for proactive retention efforts.
Data Scientist
Soundstripe
- Designed the Similar Songs recommendation engine by extracting audio features from Soundstripe's music library, decreasing user time-to-purchase by 37% and bounce rates by 53%.
- Used logistic regression to identify product features most predictive of free-to-paid conversions, enabling targeted growth strategies that optimized customer acquisition efforts.
- Applied marketing mix and Markov Chain attribution models to optimize multitouch marketing strategies, improving acquisition efficiency.
- Developed NLP-based keyword clustering models to refine search filters for sound effects, reducing user time-to-purchase by 7% and bounce rates by 46%.
- Modeled subscription health scores to identify at-risk customer segments, enabling targeted strategies to address customer churn and establishing the company’s first-ever baseline for retention improvement efforts.
- Analyzed A/B experiment outcomes to provide insights on marketing campaigns and product launches.
Experience
Similar Songs Recommendation Engine
https://www.soundstripe.com/blogs/find-music-faster-with-similar-songsSkills
Libraries/APIs
Scikit-learn, XGBoost, TensorFlow, PyTorch, Hugging Face Transformers
Tools
Git, Tableau
Languages
Python, SQL, Python 3
Paradigms
Business Intelligence (BI)
Platforms
Docker, Amazon Web Services (AWS), Google Cloud Platform (GCP), Visual Studio Code (VS Code)
Other
Statistics, Machine Learning, Deep Learning, MLFLow, Recommendation Engine, Data Analytics, Large Language Models (LLMs), Natural Language Processing (NLP), Recommendation Systems, Vector Search, Data Science, Retrieval-augmented Generation (RAG), AI Agents, FastAPI, Artificial Intelligence (AI)
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
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring