<AI_Engineer />

AboutSkillsEducationExperienceProjectsModelsPublicationsContact
Currently available for new opportunities

Youcef KADDOUR

Lead AI Engineer / Tech Lead

I build production-grade LLM & CV applications that solve real-world problems at scale.

View ResumeGet in Touch
hugging_face_logo
4
51
terminal

Technical Skills

Technologies and tools I work with

AI/ML
PyTorchExpert
TransformersExpert
LangChainAdvanced
TensorFlowAdvanced
scikit-learnExpert
Keras,Advanced
RAGAdvanced
OpenCVIntermediate
Data & Frontend
Data AnalysisExpert
NumPyExpert
PandasExpert
SparkAdvanced
SQLAdvanced
Data PipelinesAdvanced
StreamlitExpert
NextjsAdvanced
Coding
PythonExpert
JavascriptAdvanced
C++Advanced
CIntermediate
CUDAIntermediate
tritonAdvanced
JavaIntermediate
goIntermediate
MLOps
DockerExpert
KubernetesAdvanced
MLflowAdvanced
CI/CDIntermediate
Monitoring (Prometheus)Intermediate
Backend
FastAPIExpert
FlaskAdvanced
gRPCIntermediate
RESTExpert
GraphQLIntermediate
Protocol BuffersIntermediate
Cloud
AzureAdvanced
AWSIntermediate
GCPAdvanced

Education & Certifications

Academic foundation and professional certifications in AI, machine learning, and cloud technologies.

Education

M.S. Electrical Engineering

Mobile Autonomous Systems

Paris Saclay University

2019-2021

GPA: 4.0

Key Coursework:

Deep LearningComputer VisionNatural Language ProcessingReinforcement LearningStatistical Learning TheorySensor Fusion
M.S. Aerospace Engineering

Aerospace Telecommunication

Institute Of Aeronautics And Space Studies

2018-2019

GPA: 3.8

Key Coursework:

Optimization MethodsCryptography & Information SecurityDigital Signal ProcessingAdvanced Probability & StatisticsMathematical Analysis
B.S. Aerospace Engineering

Aerospace Telecommunication

Institute Of Aeronautics And Space Studies

2015-2018

GPA: 3.8

Key Coursework:

Microcontrollers & Embedded SystemsHigh-Frequency ElectronicsDigital Signal ProcessingLinear Algebra & Applied MathematicsControl Theory & Systems

Certifications

AZ-305: Azure Solutions Architect Expert

Udemy

2025
AZ-900 Bootcamp: Microsoft Azure

Udemy

2025
TOIC : 890

TOIC

2021

Professional Experience

My journey in AI and machine learning

Multi-select dropdown. Use arrow keys to navigate, Enter to select, and Escape to close.
No options selected
11/11
AI Consultant / Technical Expert
Expleo Group
2021 - Present
Paris, France
Full-time
AI StrategyExpleo Group
  • Acted as AI consultant and technical expert across aerospace, automotive, industrial, and tech sectors.
  • Delivered end-to-end AI/ML solutions including data pipelines, model development, and system integration.
  • Led discovery workshops to identify pain points, shape AI strategies, and support pre-sales proposals.
  • Produced solution architectures and technical documentation, improving proposal success rates.
  • Led R&D initiatives, set technical roadmaps, and mentored engineers and researchers.
  • Advised on LLMs, generative AI, computer vision, and ML best practices across client engagements.

Technologies Used:

PythonPyTorchTransformersFastAPIDockerKubernetesAWSMLflow
Tech Lead / Lead AI Engineer
Expleo Group
July 2024 – October 2025
Paris, France
Full-time
AI PlatformsExpleo Group
  • Led a 14-person team to build and deploy an industrial AI-agent framework for autonomous production and maintenance workflows.
  • Designed a modular multi-agent architecture (planning, control, anomaly detection, explainability) using model-based RL and rule-based reasoning.
  • Delivered internal trainings on AI, LLMs, and agentic architectures to upskill engineering teams.
  • Built a full MLOps stack (Docker, K8s, MLflow, GitHub Actions) enabling weekly retraining and zero-downtime deployment.
  • Established engineering standards and mentorship programs, increasing delivery velocity by 25%.

Technologies Used:

PyTorchLangChainHugging FaceFastAPIKubernetesDockerMLflowGitHub ActionsPostgreSQLPrometheusGCP
Tech Lead / Lead AI Engineer
Expleo Group / Airbus Helicopters
February 2023 - June 2024
Paris, France
Full-time
NLPAirbus Helicopters
  • Led development of a Text-to-SQL system achieving 91% EM accuracy for Airbus Helicopters.
  • Built a low-GPU LLM fine-tuning & inference framework (LoRA, Triton, CUDA) reducing GPU usage by 60%.
  • Defined multi-criteria evaluation metrics and built the full evaluation pipeline.
  • Standardized LLM optimization practices, documentation, and cross-team knowledge transfer.

Technologies Used:

PyTorchHugging FaceTransformersFastAPIOpenAIMySQL

Featured Projects

Open source projects and research work

Odock: Unified AI API Platform
AI Infrastructuredevelopment

A multi-tenant AI access platform with governance, budgeting, observability, and secure API key management.

GoNext.js 15TypeScriptPrisma+6
1
0
Turbo: Efficient LLM Fine-Tuning
GPU Computingdevelopment

A Python library for efficient LLM fine-tuning with adaptive GPU optimization, 4-bit loading, and LoRA enhancements.

PythonPyTorchCUDATransformers+5
1
0
Code
XLR : Inference Server
GPU Computingdevelopment

High-performance Transformer inference using Python gRPC with a C++/CUDA backend.

PythonC++CUDAgRPC+3
1
0
MedSim - Medical Training Simulator
Conversational AIdevelopment

Full-stack AI-powered medical training simulator for students and professors.

Next.js 15TypeScriptReactTailwind CSS+7
1
0
Code
FlowChat: Chatbot with Mindmap & Flow Workspace
Conversational AIdevelopment

An advanced ChatGPT-style chatbot with dual views and custom content blocks, plus an interactive mindmap for exploring conversation flows.

Next.js 15TypeScriptReactTailwind CSS+7
1
0
Alii: RAG System
NLPproduction

Production-ready RAG system for document Q&A

PythonFastAPILangChainPinecone+3
1
0
Weather Station
Embeddedproduction

Weather station using ESP32-S3, 4" RGB touchscreen (ST7701 + GT911), and BME280 sensor with LVGL UI and built-in Wi-Fi.

CLVGLESP32I2C+2
1
0
Code
Auto-AR-Drone: Autonomous Landing on Moving Platform
Robotics / CVprototype

Autonomous landing system for a Parrot AR Drone 2.0 on a moving TurtleBot platform using visual tracking, IMU fusion, Kalman filtering, and PID control.

PythonC++OpenCVKalman Filter+3
2
0
Code
AI/ML Engineer Portfolio Website
Webproduction

A modern, GitHub-inspired AI/ML engineering portfolio built with Next.js, TypeScript, and Tailwind CSS.

Next.js 15TypeScriptTailwind CSSshadcn/ui+3
1
0
CodeDemo

AI Models

Custom trained models and research contributions

Llama-3-8B-4bit

Quantized LLM for Fast Inference

Llama 3 (8B parameters) quantized to 4-bit using BitsAndBytes

Meta License

Training Data

Original Llama 3 Training Data

Training Setup

Hardware: Quantization performed on 1x NVIDIA A100 40GB
Duration: 2 hours

Performance Metrics

Perplexity-Change
+4.5%
Latency-Reduction
≈55%
Memory-Usage
≈4.65GB

Limitations

  • Quantization reduces precision in long-context reasoning
  • Not suitable for high-risk tasks requiring exact token outputs
  • Dependent on BitsAndBytes compatibility with hardware

Ethical Considerations

  • Follow Meta’s license for any commercial or derivative use
  • Must not be used to generate harmful or unauthorized content
  • Users should ensure quantized outputs remain safe and accurate
HuggingFacePaper
Llama-2-4bit

Quantized LLM for Low-Resource Inference

Llama 2 (7B/13B variant) quantized to 4-bit using BitsAndBytes

Llama 2 License

Training Data

Original Training Data

Training Setup

Hardware: Quantization performed on 1x NVIDIA A100 40GB
Duration: 1 hour 46min

Performance Metrics

Perplexity-Change
+6%
Latency-Reduction
≈60%
Memory-Usage
≈3.87GB (7B)

Limitations

  • Loss of precision in arithmetic and reasoning tasks
  • May hallucinate more under long prompts
  • Quantized models may be unstable on older GPUs

Ethical Considerations

  • Any downstream use must comply with Meta’s Llama 2 License
  • Should not be used for safety-critical or regulated decision-making
  • Quantized variants must be evaluated before deployment
HuggingFacePaper
AeroSQL-LLaMA-7B

Text-to-SQL Translation (LoRA Fine-Tuned)

LLaMA 7B (LoRA Low-Rank Adapters)

Restricted (Client-Specific)

Training Data

SpiderDomain-Specific SQL Pairs800M–70B Model Benchmarking Sets

Training Setup

Epochs: 10
Batch Size: 16
LR: 2e-4
Hardware: 1/2x NVIDIA A100 80GB
Duration: 72 hours

Performance Metrics

Exact-Match
0.91
Execution-Accuracy
0.88
Schema-Generalization
0.85

Limitations

  • Requires careful prompt engineering
  • May hallucinate table/column names
  • Still struggles with long schema contexts

Ethical Considerations

  • SQL outputs must pass validation before execution
  • Aviation operational data must be kept private
  • Should not be used for mission-critical queries without review
AeroSQL-Mistral-7B

Text-to-SQL Translation (LoRA Fine-Tuned)

Mistral 7B (Dense Transformer, LoRA)

Restricted (Client-Specific)

Training Data

SpiderAirbus Helicopters SQL LogsSynthetic Query Expansions

Training Setup

Epochs: 9
Batch Size: 16
LR: 1.5e-4
Hardware: 1/2x NVIDIA A100 80GB
Duration: 53 hours

Performance Metrics

Exact-Match
0.92
Execution-Accuracy
0.89
Reduced-Hallucinations
0.90

Limitations

  • Still dependent on schema clarity
  • Performance varies on extremely long SQL trees
  • Fails on rare domain-specific operators

Ethical Considerations

  • Human verification required for SQL execution
  • Training logs must be anonymized
  • High accuracy does not eliminate hallucination risks

Publications & Talks

Research papers, blog posts, and conference presentations

Featured Publications

paper1 citations
Location and Landing of an Autonomous Drone on a Mobile Platform
Youcef KADDOUR
Paris-Saclay University • January 2020 – June 2020

Development of a real-time detection, monitoring, and landing system enabling an autonomous drone to land on a moving platform. Work included updating and extending a Python library for Parrot AR-Drone 2.0 control, implementing a computer-vision detection pipeline using ArUco markers, Kalman filtering, and GOTURN tracking, and creating a real-time trajectory planning and correction system.

Autonomous DronesComputer VisionArucoKalman FilterOpenCVRobotics
PaperVideo
paper1 citations
Evaluation of Visual Monitoring and Tracking Based on Deep Learning Algorithms
Youcef KADDOUR
Paris-Saclay University • September 2020 – February 2021

Comprehensive study and categorization of state-of-the-art visual tracking and monitoring methods based on Deep Learning, including CNN, RNN, SNN, YOLO-v3, and DRLT. The project included performance evaluations across multiple benchmarks with respect to robustness, computational requirements, memory usage, and adaptability across various scenarios. Key limitations of each method were identified to support method selection for real-world applications.

Deep LearningVisual TrackingComputer VisionDRLTYOLONeural Networks
Paper

Get In Touch

Let's discuss opportunities and collaborations

Contact Information

Email

youcef.kaddour.pro@gmail.com

Location

Cannes, France

Book a meeting

Availability

Currently available for new opportunities

Connect

Send a Message

Blog

Coming soon...

Built by Youcef KADDOUR

Last updated: 11/17/2025

AboutProjectsContactPrivacy