Hello!👋🏾

I'm Thierno Daouda

Data & AI Engineer

About me

Computer Science Engineer with a strong foundation in algorithms, distributed systems, and applied mathematics. I focus on Artificial Intelligence and Data Engineering, with hands-on experience building scalable systems, modern data platforms, and AI-driven applications.

I enjoy combining theoretical foundations with hands-on experimentation to understand how modern AI systems learn, reason, and process information. I’m particularly interested in challenging technical and research-oriented projects that turn ideas into practical solutions.

Programming Mathematics Data Science Data Engineering Machine Learning Artificial Intelligence Research

Experience

Junior Data & AI Engineer

Orange

Jan. 2026 - Jun. 2026

Designed and implemented an end-to-end Data Lakehouse architecture built on a Medallion pattern, integrating scalable object storage, distributed query engines, streaming ingestion, semantic transformations, and workflow orchestration.
Built an open-source data platform based on MinIO, Trino, Iceberg, dbt, Airflow, Kafka, and Flink, covering ingestion, storage, transformation, and observability layers.
Developed a custom multi-agent orchestration framework, based on a four-agent ReAct paradigm, for pipeline supervision, anomaly detection, and autonomous workflow optimization.

AI Engineering Intern

Orange

Jul. 2025 - Dec. 2025

Designed and deployed a Retrieval-Augmented Generation (RAG) engine for semantic search enabling natural-language querying over enterprise content.
Developed an end-to-end document ingestion and indexing pipeline covering parsing, chunking, embedding, vector storage, retrieval, and automated evaluation.
Improved retrieval performance through optimized chunking strategies, hybrid retrieval techniques, and systematic LLM-based evaluation of answer relevance and faithfulness.

Software Engineering Intern

Orange

Sep. 2024 - Dec. 2024

Designed and developed a data lineage web application for visualizing database dependencies and metadata across enterprise systems.
Modeled enterprise data relationships as a property graph in Neo4j, strengthening data traceability and governance capabilities.
Implemented Django backend services and Cypher graph queries to track data flows and transformations across systems.

Projects

Ilam: AI-Augmented Data Lakehouse Platform

Open source Data Lakehouse platform unifying high-performance storage, real-time stream processing, semantic transformation, and intelligent orchestration through a Medallion architecture, a LangGraph-powered multi-agent layer, and a full-stack observability-driven portal

DockerKubernetesMinIOTrinodbtIcebergAirflowFlinkKafkaGreat ExpectationsPrometheusGrafanaLangGraphFastAPIReact

Séet: Semantic Enterprise Search Engine

Multimodal RAG system for complex financial report analysis and question answering, combining dense semantic retrieval with vision-language understanding to extract insights from text, tables, and charts embedded in enterprise documents, indexed through ChromaDB and OpenSearch and evaluated with RAGAS

DockerPythonLLMsChromaDBClaude VisionBGE-M3OpenSearchRAGAS

Graphlens: Data Lineage Visualization App

Django Web application for managing and visualizing data, table metadata and lineage relationships, modeling entities and dependencies as a property graph in Neo4j and traversing them with Cypher queries to surface upstream and downstream impact across complex data pipelines

DjangoNeo4jNeomodelCypher

Artify: Neural Style Transfer Implementation

PyTorch implementation of Neural Style Transfer (Gatys et al. 2016), extracting hierarchical content and style representations from a pretrained VGG19 network, computing style loss via Gram matrices, and iteratively optimizing the generated image with L-BFGS, wrapped in a Streamlit interface for interactive experimentation

DockerPyTorchVGG19Gram MatrixL-BFGSStreamlit

Tontuma: Visual Question Answering Platform

VQA app powered by the Visual-and-Language Transformer (ViLT), jointly encoding image patches and natural language questions through a unified transformer architecture to deliver instant answers grounded in visual content, with no separate object-detection pipeline required

PythonViLTStreamlitVQA

Education

Engineering Degree in Computer Science

École Polytechnique de Thiès

Jan. 2022 - Jun. 2026

Intensive foundation in mathematics, probability, statistics, and engineering sciences to prepare for the engineering cycle
Completed advanced coursework in computer science and programming, building the groundwork for later specialization in Data & AI
Activities: Junior Polytech Competition, student journalism, Fablab Manager, academic quiz competitions, athletics

High School Diploma

Lycée Scientifique d'Excellence de Diourbel

Oct. 2018 - Jul. 2021

Selective national excellence program focused on advanced mathematics and physics for top-performing students
Recipient of the National History Prize at the Concours Général Sénégalais
Activities: Mathematics & Physics Club, academic quiz competitions

Other things!

Outside of coding, I enjoy reading, watching films, and learning about history & science. I always enjoy discovering new ideas, exploring different perspectives, and learning things that challenge the way I think.