I'm an AI Engineer who specializes in leveraging large language models for business applications and building & deploying multi-agent and RAG systems.…

Kparobor Akpomiemie

I'm an AI Engineer who specializes in leveraging large language models for business applications and building & deploying multi-agent and RAG systems.…

Available to hire

I’m an AI Engineer who specializes in leveraging large language models for business applications and building & deploying multi-agent and RAG systems.

Skills

Experience Level

Data Science
Expert
Python
Expert
TensorFlow
Expert
Github
Expert
PyTorch
Expert
Data Visualization
Expert
Docker
Expert
R
Expert
AWS
Expert
SQL
Expert
MongoDB
Expert
MySQL
Expert
Google Cloud
Expert
PostgreSQL
Expert
Power BI
Expert
Keras
Expert
Kubernetes
Expert
Azure
Expert
Scala
Expert
Java
Expert
See more

Language

English
Fluent
French
Beginner

Work Experience

GenAI Engineer at ENGIE
July 1, 2025 - October 1, 2025
Developed an LLM-powered classification pipeline using a fine-tuned Camembert-base model to categorize customer complaints into groups for the marketing department, reducing manual processing time by over 90%. Designed and deployed an LLM-powered churn prediction model leveraging XLM-Roberta and nlptown/bert-base-multilingual-uncased-sentiment models for churn predictions and sentiment analysis, achieving 81% accuracy in identifying potential customer churn.
ML Engineer (Time-Series Specialist) at Prosol
March 1, 2025 - October 1, 2025
Developed Random Forest and Prophet models for demand forecasting across 20 stores and 3,445 products, optimizing supply-demand balance with engineered features, achieving a 44% MAE reduction for 7-day SKU-level forecasts. Designed a real-time Tableau dashboard using SQL to visualize 7-day demand forecasts, enabling cross-functional teams to balance supply (inventory) and demand (purchases), reducing supply chain costs by 20%. Identified Prophet’s limitations in volatile retail demand and recommended Temporal Fusion Transformers for scalable forecasting, enhancing supply-demand alignment and future-proofing operations.
Data Scientist at ItaraShop
June 1, 2024 - October 1, 2025
Performed K-means clustering to identify high-value customers to drive personalized offers, increasing customer targeting by 30%. Developed forecasting models using an ensemble (XGBoost, Random Forest, LightGBM – 89% accuracy) and Facebook Prophet (87% accuracy) to predict products to sell over a 7-day period, increasing product stocking by 25%. Conducted causal inference analysis using propensity score matching to evaluate product recommendations, identifying a 15% causal increase in cart additions and 10% uplift in repeat purchases.
Head of Strategy and Product at ItaraShop
December 1, 2021 - October 1, 2025
Led the launch of ItaraShop app, including a product recommendation system which generated $300,000 in revenues in 24 months. Led multi-channel customer acquisition strategy resulting in 50% growth in customer acquisition. Onboarded 200+ African brands through target outreach, expanding platform inventory to over 1000 unique products improving customer satisfaction by over 50%.
LangGraph E-commerce Analytics Agent at LangGraph
September 1, 2025 - September 30, 2025
A custom ReAct (Reason-Act) multi-agent AI system designed to analyze natural language queries over the publicly available BigQuery dataset bigquery-public-data.thelook_ecommerce. The project leverages Google’s Gemini 1.5 Flash LLM and LangGraph to translate complex queries into validated SQL, enabling real-time, explainable e-commerce analytics. Engineered a robust multi-agent state machine with persistent conversational memory, enabling multi-turn query sessions and precise agent delegation. Developed advanced chain-of-thought and reflection nodes to validate and iteratively refine SQL queries dynamically, ensuring over 99% SQL validity in production. Built modular, specialized sub-agents (segmentation, trends, geo, product) that utilize live BigQuery data to generate rich business insights with sub-10-second response times.

Education

Master of Science in Data Science & AI Strategy at Emlyon Business School, Paris, France
September 2, 2024 - October 1, 2025

Qualifications

AWS Certified Cloud Practitioner
April 5, 2024 - April 6, 2028
Oracle Cloud Infrastructure 2025 Certified Foundations Associate
January 11, 2030 - October 1, 2025
Oracle Cloud Infrastructure 2025 Certified AI Foundations Associate
January 11, 2030 - October 1, 2025
Oracle Cloud Infrastructure 2025 Certified AI Foundations Professional
January 11, 2030 - October 1, 2025
Hugging Face AI Agent Course
January 11, 2030 - October 1, 2025
LinkedIn: Master Retrieval Augmented Generation
January 11, 2030 - October 1, 2025
LinkedIn: Fine-tuning for LLMs: from Beginner to Advanced
January 11, 2030 - October 1, 2025
LinkedIn: Agentic AI Fundamentals: Architectures, Frameworks, and Applications
January 11, 2030 - October 1, 2025
AI Policy Clinic certification
January 11, 2030 - October 1, 2025

Industry Experience

Software & Internet, Retail, Other
    paper Custom LangGraph agent for e-commerce analysis with BigQuery and Gemini 1.5 Flash

    LangGraph E-commerce Analytics Agent

    An E-commerce Analytics Agent powered by LangGraph and Google’s Gemini 1.5 Flash LLM.Google Gemini 1.5 Flash is particularly well-suited for SQL generation tasks due to its strong reasoning capabilities combined with a large context window, enabling it to understand complex database schemas and user intents in detail. Its architecture supports the generation of precise, schema-compliant SQL queries by leveraging chain-of-thought prompting and iterative refinement, which reduces ambiguity and errors in the generated code. This makes Gemini 1.5 Flash ideal for translating natural language questions into accurate SQL, ensuring reliable execution on large datasets like bigquery-public-data.thelook_ecommerce, while maintaining low latency and cost-effective inference for real-time analytics applications.

    Query the bigquery-public-data.thelook_ecommerce dataset using natural language and receive actionable, business-ready analytics in seconds — no SQL skills required.

    View the project here: https://www.twine.net/signin

Hire a AI Engineer

We have the best ai engineer experts on Twine. Hire a ai engineer in Paris today.