Bridging AI research and real-world applications through intelligent document understanding.
I am an AI Research Engineer and PhD Researcher focused on the extraction and analysis of unstructured data within Electronic Document Management Systems (EDMS).
My work combines Natural Language Processing, Information Retrieval, Large Language Models, and Knowledge Management to transform enterprise documents into structured, actionable knowledge.
Beyond research, I design and deploy production-grade AI systems involving Retrieval-Augmented Generation (RAG), MLOps, LLMOps, and Agentic AI architectures.
My doctoral research focuses on developing intelligent systems that extract, structure, and analyze information from large-scale unstructured document repositories.
- Natural Language Processing (NLP)
- Information Extraction
- Retrieval-Augmented Generation (RAG)
- Semantic Search and Retrieval
- Knowledge Graph Construction
- Large Language Models for Document Understanding
- Enterprise Knowledge Management
- Intelligent Document Processing (IDP)
- Extract structured information from unstructured documents
- Improve document classification and semantic indexing
- Develop intelligent enterprise search systems
- Leverage LLMs for knowledge extraction and reasoning
- Design scalable document intelligence pipelines
- Advanced Hierarchical Classification Approach for Document Categorization
- Authors: Mamadou Alpha Hawa Balde; Pirlouit Dumez; Meriam Belgaroui; Guillaume Prevost; Salah Zidi
- Conference: IEEE Afro-Mediterranean Conference on Artificial Intelligence (AMCAI), 2026
- DOI: 10.1109/AMCAI66110.2025.11474373
Extraction of structured information from enterprise document repositories using NLP and LLMs.
Building retrieval-augmented generation systems for contextual and semantic knowledge access.
Transforming unstructured text into structured knowledge representations.
Designing pipelines for entity extraction, classification, and document understanding.
✓ Retrieval-Augmented Generation (RAG) Systems
✓ Production AI Applications
✓ LLM Evaluation Pipelines
✓ Intelligent Agents
✓ Scalable Data Platforms
✓ Cloud-Based ML Infrastructure
A production-grade Retrieval-Augmented Generation platform enabling intelligent knowledge retrieval and contextual question answering.
Key Features
- Vector Search
- Semantic Retrieval
- Multi-Document Processing
- LLM Integration
- Evaluation Pipeline
Scalable ETL and orchestration framework designed for modern data workflows.
Key Features
- Automated Data Validation
- Data Quality Monitoring
- Cloud Storage Integration
- Workflow Automation
End-to-end machine learning lifecycle management including training, deployment, and monitoring.
Key Features
- Experiment Tracking
- Model Registry
- Automated Deployment
- Continuous Monitoring
Production-ready framework for deploying, evaluating, and monitoring LLM-powered applications.
Key Features
- Prompt Management
- Cost Tracking
- Evaluation Workflows
- Performance Monitoring
- Agentic AI Systems
- Multi-Agent Architectures
- Advanced LLMOps
- AI Evaluation Frameworks
- Production AI Infrastructure
- Build scalable AI agents for real-world systems
- Publish research in NLP and document intelligence
- Contribute to open-source AI projects
- Advance LLMOps and Agentic AI engineering
- Develop enterprise-grade AI solutions
📧 Email: baldehalfahim@gmail.com 💼 LinkedIn: Coming Soon 🌐 Portfolio: Coming Soon