Benjamin Jodoin

Senior Data Analyst with 7+ years of experience building enterprise-scale AI platforms leveraging Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), transformer architectures and Computer Vision. Skilled in developing end-to-end AI solutions including data ingestion, embedding pipelines, vector databases, prompt engineering, model fine-tuning, inference optimization and cloud deployment. Demonstrated success leading engineering teams, delivering scalable production systems and improving AI performance through advanced optimization techniques.

Strong expertise in Python, PyTorch, TensorFlow, LangChain, Hugging Face Transformers, FAISS, SQL, Docker, Kubernetes and AWS/Azure/GCP.

GenAI & LLM Engineer - Dust

(2025-09 - 2026-06)

• Led development of a GenAI-powered enterprise chatbot using LLMs and RAG architecture, reducing manual support workload by 45%
• Designed and implemented RAG pipelines with FAISS, improving answer relevance by 35% and reducing hallucinations
• Fine-tuned transformer models using domain-specific datasets, increasing response accuracy by 25%
• Built scalable ETL pipelines processing 10M+ records for real-time AI inference
• Optimized inference latency by 30% through model pruning and efficient embedding strategies
• Managed and mentored a team of 6 data scientists and ML engineers in Agile delivery cycles

Team Lead & Machine learning Engineer - Cognizant

(2021-08 - 2025-07)

• Developed LLM-based semantic search engine handling 1M+ queries/day with sub-second latency
• Built NLP pipelines for text classification and entity recognition using transformer models (BERT/GPT)
• Designed computer vision models for object detection achieving ~92% accuracy using CNN architectures
• Improved model performance by 20% through hyperparameter tuning and data augmentation techniques
• Deployed containerized ML models using Docker and AWS, ensuring scalable and fault-tolerant systems
• Collaborated with data labeling teams to improve dataset quality, increasing model robustness

Python Developer - Scale AI

(2018-07 - 2021-06)

• Developed machine learning models for predictive analytics, forecasting, and business intelligence applications
• Built data pipelines and automation scripts using Python and SQL for large-scale data processing
• Implemented NLP solutions for sentiment analysis, document classification, and text mining
• Designed interactive dashboards and reporting tools to support data-driven decision-making
• Contributed to Agile development cycles, delivering AI-driven client solutions within tight deadlines

BS - computer Science - EURECOM (2012-08 - 2017-12)