Description

Data Scientist / LLM / Generative AI Engineer with 8+ years of experience in

GenAI, LLMs, NLP, and Deep Learning, specializing in RAG pipelines,

prompt engineering, AI agents, and production-grade AI systems.

Expert in LLM fine-tuning (LoRA/QLoRA), semantic search, embed-

dings, and LLM evaluation, with strong experience in end-to-end

ML/LLM pipelines from PoC to scalable deployment on Microsoft

Azure, Amazon Web Services, and Google Cloud Platform.

Proven track record in risk modeling, fraud detection, and customer

analytics, focusing on scalable system design, AI safety, and busi-

ness impact.

KEY SKILLS

• Generative AI & LLMs: Prompt Engineering (Zero-shot, Few-shot,

Chain-of-Thought), Retrieval-Augmented Generation (RAG), LLM

Fine-tuning (LoRA, QLoRA), RLHF, Context Optimization, Hallucina-

tion Mitigation, LLM Evaluation (RAGAS, BLEU, ROUGE, Faithfulness

Metrics)

• Frameworks & Libraries: LangChain, LlamaIndex, Hugging Face

Transformers, PyTorch, TensorFlow, Scikit-learn

• Vector Databases & Search: FAISS, Pinecone, Weaviate, Semantic

Search, Embedding Models, Approximate Nearest Neighbor (ANN)

Programming & Development: Python, SQL, NumPy, Pandas, REST

APIs, FastAPI, Async Programming

Cloud & MLOps: Microsoft Azure (Azure OpenAI, Azure ML), Amazon

Web Services (SageMaker, Bedrock), Google Cloud Platform (Vertex

AI), Docker, Kubernetes, CI/CD, MLflow, Model Deployment & Moni-

toring

System Design & Architecture: End-to-End LLM Application De-

sign, Scalable RAG Pipelines, Prompt Pipelines, Context Manage-

ment, Cost Optimization, API Integration

AI Safety & Governance: Prompt Injection Prevention, Data Privacy,

Content Filtering, Responsible AI Practices

Languages

English
Fluent
German
Conversational
French
Basic

Workplace preferences

Can work on-site

Paris (up to 50km)

SFN Germany
Senior AI engineer
April 2025 - Today (1 year and 3 months)
Designed and developed a production-grade Retrieval-
Augmented Generation (RAG) conversational AI system for
processing sports data, guidelines, and real-time news tailored
for B2B applications.
• Engineered scalable LLM pipelines using LangChain, FastAPI, and
vector databases to enable semantic search, context-aware re-
sponses, and high retrieval accuracy, significantly improving re-
sponse relevance and system performance.
Hawkscode
Senior Data Scientist
January 2017 - September 2020 (3 years and 8 months)
Developed credit risk and fraud detection models using Python,
Spark, and Amazon Web Services, improving accuracy by 20%.

Built scalable ML pipelines with focus on reproducibility, explain-
ability, and production deployment.

Designed customer segmentation models and executed A/B test-
ing for targeted marketing optimization.

• Implemented data drift detection and automated retraining
pipelines to ensure model stability.

• Developed batch and real-time scoring systems using Spark, SQL,
and Databricks and mentored junior team members.
Applied Data Finance
Data Scientist
BANKING AND INSURANCE
August 2015 - January 2017 (1 year and 5 months)
Built and deployed risk and fraud detection models using transac-
tional data exceeding 50M records with Scikit-learn, Pandas, AWS
and MySQL.
• Implemented CI/CD pipelines and data-driven experimentation for
scalable production ML
Python MySQL Machine learning Data science Spark