MARTINS EDEGBE
I am a Data Scientist and Machine Learning Engineer with a strong background in predictive analytics, Generative AI, and production-grade machine learning systems.
About
I am a Data Scientist and Machine Learning Engineer with a strong background in predictive analytics, Generative AI, and production-grade machine learning systems. I hold a Master of Applied Science in Mechanical Engineering from the University of Windsor, where my research focused on computational modelling, fluid dynamics, and machine learning applications.
Over the years, I have built and deployed end-to-end AI and analytics solutions across customer intelligence, forecasting, anomaly detection, sentiment analysis, and business automation. My work spans traditional machine learning, statistical modeling, Retrieval-Augmented Generation (RAG), LLM orchestration, and agentic AI systems using tools such as Python, SQL, Snowflake, Databricks, LangChain, LangGraph, FastAPI, and cloud platforms like AWS and GCP.
I have developed churn prediction models, customer lifetime value forecasting systems, CRM analytics pipelines, and AI-powered assistants that transformed complex structured and unstructured data into actionable business insights. One of my most notable projects was a production-grade RAG-based AI assistant for a restaurant business, designed to automate customer support by retrieving grounded information from operational documents, significantly reducing repetitive support workload and improving response speed.
What makes me stand out is my ability to bridge technical depth with business value. I do not just build models, I design scalable AI systems, optimize workflows, monitor performance, and translate ambiguous business challenges into measurable solutions. My experience spans analytics, intelligent automation, LLM applications, and enterprise AI systems, with a strong focus on reliability, governance, and practical impact.
- 7 years' experience
- Joined July 2026
Industry experience
- Financial Services
- Telecommunications
- Software & Internet
- Professional Services
- Media & Entertainment
Work experience· 4 roles
Applied GenAI techniques (LLM-driven sentiment analysis, zero-shot classification, summarization) using OpenAI APIs, prompt engineering, and embeddings-based retrieval to categorize and prioritize multi-channel customer feedback; designed and deployed agentic AI and RAG systems using LangChain and LangGraph for context-aware question answering over unstructured knowledge bases; implemented information retrieval pipelines to improve retrieval precision/recall and response quality for RAG applications; built and extended Model Context Protocol (MCP) servers and clients for interoperable AI integrations across internal systems and third-party tools; developed scalable data pipelines in Python and SQL to support automated training and inference workflows; implemented rollback strategies for ML and data pipelines; enhanced RAG systems with Snowflake Cortex embedding and retrieval; fine-tuned LoRA-based models; containerized LLM APIs; built production AI/ML workflows; deployed on Google Clou
Applied computational sciences foundations to analyze large spatiotemporal datasets and develop reliable predictive models for complex unsteady system dynamics; conducted advanced time series analysis and spectral decomposition techniques (including Spectral Proper Orthogonal Decomposition); developed and trained ML/DL models (CNNs) for predictive modeling on high-dimensional temporal data; evaluated model performance using cross-validation, residual analysis, and statistical error metrics; communicated findings via technical reports, conference presentations, and manuscripts.
Developed and deployed churn prediction models; built metrics to track support volume drivers and operational bottlenecks; engineered customer usage features from CDRs and billing logs; integrated Zuora subscription/billing data to monitor MRR/ARR and lifecycle analytics; created Power BI dashboards for executives; conducted A/B testing on data plans and incentives; performed time series analysis for usage and network activity; developed scalable ETL pipelines in Databricks (PySpark and SQL) with cloud storage and CI/CD integration; built SQL-based analytics pipelines with BigQuery and Snowflake.
Education
Certifications and awards
Hire MARTINS EDEGBE today
Post a job and invite MARTINS, or contact them directly. No hidden commission.