A vision model that scores photographs on composition rules (rule of thirds, leading lines, depth of field), with explainability via Grad-CAM. Trained on a personal dataset I annotated from my own photographs.
PythonPyTorchFastAPIReactGrad-CAM
William Kau.
AI & Software Engineer.
M2 engineering student at ESILV, building AI solutions around LLMs and synthetic data. Currently a Software Engineer intern at Aubay Solutec, with prior experience at BPCE SI and Manaos.
- Based in
- Paris, FR
- Studying
- ESILV - M2
- Now
- Aubay Solutec
- Focus
- LLMs · Data
◆Python
◆SQL
◆C#
◆Java
◆FastAPI
◆Angular
◆React
◆Next.js
◆PyTorch
◆TensorFlow
◆Hugging Face
◆LangChain
◆SDV
◆SynthCity
◆PostgreSQL
◆Oracle SQL
◆MinIO
◆GCP
◆Docker
◆Git
◆Power BI
◆Streamlit
◆Python
◆SQL
◆C#
◆Java
◆FastAPI
◆Angular
◆React
◆Next.js
◆PyTorch
◆TensorFlow
◆Hugging Face
◆LangChain
◆SDV
◆SynthCity
◆PostgreSQL
◆Oracle SQL
◆MinIO
◆GCP
◆Docker
◆Git
◆Power BI
◆Streamlit
About
Engineer driven by data and curiosity.
I'm William, a final-year Data Science & AI engineering student at ESILV Paris La Défense, graduating in June 2026. I'm currently a Software Engineer intern at Aubay Solutec in Lyon, where I train generative models (CTGAN, TVAE, DDPM) to produce synthetic data faithful to real-world distributions, and ship them inside a FastAPI / Angular micro-services app.
My work lives at the intersection of machine learning, LLM systems and production engineering: generative models, RAG pipelines with traceable citations, and agentic loops, deployed with FastAPI, Docker and GCP. I care about AI systems people can actually trust, inspect, and run.
Outside the screen, I'm a setter on a competitive volleyball team, I run trails, climb, and shoot photographs. More often than I expected, these interests seep back into the way I think about code.
5+
Years coding
3
Internships
1st
ESILV PI2 prize
975
TOEIC / 990
Experience
Where I've been.
Three internships at the intersection of data engineering, AI and product, each one a step deeper into building things that ship.
- Feb 2026 - Present
Software Engineer - Internship @ Aubay Solutec
Lyon, FranceSelecting and tuning 4 generative algorithms (CTGAN, TVAE, ARGN, DDPM) to produce synthetic data faithful to real-world distributions. Designed a persistence pipeline for trained models and generated datasets (MinIO + PostgreSQL), and built the fullstack micro-services app (FastAPI / Angular) with JWT authentication (OAuth2, bcrypt).
FastAPIAngularPyTorchCTGANTVAEDDPMMinIOPostgreSQLJWT - Apr 2025 - Aug 2025
Data Analyst - Internship @ BPCE SI
Paris 13e, FranceOptimised SQL queries on Oracle databases, cutting execution time of recurring processes by up to 60%. Industrialised BI reports and contributed to the migration of the data heritage to Google Cloud Platform.
Oracle SQLPower BIGCP - Sept 2024 - Apr 2025
Fullstack Developer - School Project @ MANAOS - BNP Paribas
Paris 8e, France★ 1st prize - ESILV PI2 2024-2025Built an ESG data management application in Python / Streamlit (team of 6) with an integrated open-source LLM (Hugging Face) to query the data in natural language.
PythonStreamlitHugging FaceLLMESG
Selected work
Projects I'm proud of.
A mix of school, internship and personal work, usually somewhere between AI research and shipping software.
A retrieval-augmented system for querying financial and ESG reports in natural language. 100% local stack with semantic retrieval, source citations and a hallucination-detection layer (NLI + grounding + self-consistency).
PythonOllamaChromaFastAPIStreamlit
A full-stack app where users ask questions in natural language and an Anthropic-powered agent generates, validates and executes SQL against a database. The UI surfaces the final SQL, validation status, the agent's retry attempts and the result table.
PythonAnthropicSQLAgents
2026
Synthetic Data Generation Platform
Generative models, productionised
Ongoing at Aubay Solutec: a fullstack micro-services platform that picks and tunes generative algorithms (GAN, VAE, ARGN, DDPM) based on the input dataset. FastAPI + Angular, MinIO and PostgreSQL on Nexus.
FastAPIAngularGANVAEDDPMPostgreSQL
Off-screen
What keeps me sharp.
The hours I spend away from a keyboard, and why they end up shaping my engineering work more than I expected.
Volleyball
Setter for a competitive amateur team in Lognes. The position taught me a lot about anticipating, reading patterns, and making quick calls under pressure.
Running & Trail
Long runs and trail outings: the rhythm I rely on to think through hard problems, away from any screen.
Climbing
On the wall I get to optimise something different: balance, route reading, body tension. Debugging a route is a lot like debugging code.