Applied AI & Solution Engineering | LLMs · RAG · Machine Learning · APIs
I'm a graduate engineer (M.Eng. in Data & Computer Science) from École des Mines de Saint-Étienne, with a strong foundation in mathematics and physics from two years of intensive preparatory classes at Lycée Saint-Louis (Paris).
I’m passionate about turning complex business challenges into practical AI solutions that deliver measurable value.
📍 Paris, France · 🎯 Open to Solution Engineer / AI Solution Engineer opportunities
- Applied AI & LLM systems: retrieval-augmented generation (RAG), agentic workflows, NLP, ML and deep learning.
- Solutions & integrations: Python, JavaScript / Node.js, FastAPI, REST APIs, PostgreSQL, SQL, Docker.
- Data engineering & analytics: ETL, Snowflake, data modeling, forecasting, dashboards and automation.
- Engineering foundations: algorithms, optimization, systems programming, networking, cybersecurity and embedded systems.
During my final engineering internship at SAP, I worked on applied AI solutions for B2B sales teams, combining business discovery, software engineering and user feedback.
- AI-powered CRM — data quality & enrichment prototype: designed and developed a CRM intelligence platform to consolidate, verify and enrich business account data, with a Python/FastAPI backend, PostgreSQL and machine-learning-assisted data-quality workflows.
- AI Sales & Account Planning Agent: built an LLM-powered assistant to help Business Development Representatives (BDRs) and Account Executives prepare account plans and meeting briefs using relevant CRM context.
These were company projects. The source code, internal data, architecture details and application screenshots are not publicly available due to confidentiality obligations. The descriptions here intentionally focus on my responsibilities and the general capabilities rather than proprietary implementation.
📖 Read the public case study (confidentiality-safe overview).
I built a synthetic B2B presales evaluation extension on top of Microsoft's Azure Search OpenAI Demo (MIT-licensed upstream). The underlying Azure RAG application and infrastructure are Microsoft's work; my contribution is the original business scenario corpus, 12 structured regression cases, a Python evaluation harness with optional API integration, offline tests, a privacy-conscious HTML reporting tool and complete solution-engineering documentation.
This project demonstrates how I approach customer-facing AI solutions: identify the business question, ground answers in trusted data, test for unsupported claims, explain limitations and present results. The fixture-based demonstration works without Azure; a live cloud deployment requires Azure resources and has not been claimed as completed.
Explore the fork and contribution → · Technical walkthrough →
Eight projects reflecting my interests in applied AI, solution engineering, optimization, software development and systems design.
| Project | What it demonstrates |
|---|---|
| RFID Localization — Machine Learning | UHF RFID localization from multi-antenna measurements, feature engineering, modeling and experimental evaluation. |
| Minimum Inventory Optimization | Operations research and Python-based optimization applied to inventory management. |
| Azure RAG Presales Evaluation — Open-source Extension | Synthetic B2B scenarios, Python evaluation harness, answer-quality checks and reports built on Microsoft's Azure RAG sample. |
| ECG Classification — Machine Learning | ECG preprocessing, PCA, clustering and classification benchmarks with multiple machine-learning models. |
| Supply Chain Optimization — Java | Scheduling optimization and heuristic approaches including BIBA, VND and LAHC. |
| Tennis REST API — NestJS | Backend engineering, CRUD APIs and integration with a third-party sports data API. |
| Multiplayer Chess Server — C | TCP client–server architecture, concurrent games, chess-rule validation and automated tests. |
| ASCON-128 — SystemVerilog Cryptographic Hardware | Hardware design in SystemVerilog, cryptographic architecture and RTL implementation with a simulation testbench. |
Across academic, personal and open-source work, I've explored diverse fields: reinforcement learning, time-series machine learning, network security, embedded robotics, IoT prototyping and interactive web applications.
- SAP — Sales AI Engineering: built and demonstrated AI-assisted CRM and sales enablement solutions, from business requirements to implementation and user feedback.
- Generali France — Data Engineering: data pipelines, automation, validation and analytical data models.
- SNCF Group — Data Engineering: REST API integrations, geospatial datasets and data reconciliation.
My professional work includes confidential company projects. Their repositories, internal datasets and proprietary implementation details are intentionally not published here.
Understand the use case → design the architecture → build a prototype → evaluate it → explain the value.
I particularly enjoy roles at the intersection of customer/business needs, software engineering and AI, where technical depth and clear communication matter equally.
Languages: French (native) · English (TOEIC 965/990) · Spanish (advanced).