Achraf Habib

Achraf Habib

|

Building production-grade AI systems with Multi-Agent architectures, LLMs, and cloud-native solutions

Get In Touch Download CV

About Me

I'm an AI Engineer based in Paris, currently at Socotec Monitoring, where I architect and deploy production-grade, LLM-powered multi-agent systems using LangGraph and AWS Bedrock.

With a strong foundation from INSAT's Software & AI Engineering program and an ongoing MSc in Generative AI at AIvancity School (Eduniversal Innovation Award), I bring deep expertise across the full AI stack — from model training and fine-tuning to cloud deployment and MLOps pipelines.

My work spans multi-agent orchestration, computer vision, NLP, and scalable cloud architectures. I thrive on turning cutting-edge AI research into reliable, production-ready systems.

1.5+ Years Experience
4 Companies
5+ Production Systems
2 Degrees

Professional Experience

SOCOTEC 1 yr 3 mos

AI Engineer

Apprenticeship

Paris, France — On-site
Nov 2025 — Present
  • Developing and deploying production-ready, LLM-powered agents using the LangGraph Python SDK and AWS Bedrock
  • Designing scalable AI solutions with emphasis on reliability, performance optimization, and production readiness
LangGraphAWS BedrockGitHubProject ManagementPython

AI Engineer

End-of-Study Internship — SOCOTEC

Paris, France — On-site
Mar 2025 — Nov 2025
  • Designed and implemented a production-grade, LLM-driven agent using LangGraph and AWS Bedrock
  • Developed an OCR-based agent leveraging AWS Textract and advanced LLMs for efficient image-to-text extraction, integrated with structured database querying for seamless data processing
  • Built and deployed a YOLO-based object detection model on AWS SageMaker for real-time inference
  • Implemented a complex Kubernetes architecture to enable automated data annotation workflows, ensuring scalability, continuous model updates, and optimized data processing
LangGraphAWS BedrockAWS TextractAWS SageMakerAWS EKSAWS RDSYOLOKubernetesPython

Part-Time AI Engineer

Lightray Technologies

Cyprus — Remote
Sep 2024 — Dec 2024
  • Contributed to a large-scale financial market prediction system leveraging LLaMA 3.1 405B to replicate institutional investor decision-making workflows
  • Led data collection and augmentation efforts, expanding training corpus by 25% through advanced web scraping methodologies
  • Optimized fine-tuning pipeline using Snowflake framework, achieving 30% reduction in training time through multi-GPU architecture
LLaMA 3.1 405BLoRATensorRT-LLMTritonMulti-GPUSnowflake

MLOps Intern

Applicare.AI

Montreal, Canada — Remote
Jun 2024 — Sep 2024
  • Optimized LSTM, GRU, and Transformer architectures on the MIMIC-III clinical dataset, achieving 25% improvement in prediction accuracy
  • Developed and containerized a real-time patient outcome prediction application using Docker, deployed on Azure with comprehensive MLOps pipeline integration via MLflow
  • Enhanced model reproducibility by 40% through end-to-end pipeline automation with GitHub Actions
MLflowDockerAzureGitHub ActionsStreamlitMIMIC-III

Generative AI Intern

VERMEG

Tunis, Tunisia
Jul 2023 — Aug 2023
  • Developed an intelligent chatbot leveraging LangChain's RAG architecture, achieving 25% improvement in response accuracy for Vermeg's proprietary Palmera framework
  • Enhanced document retrieval efficiency by 30% through implementation of vector embeddings and optimized similarity search with ChromaDB
  • Designed and deployed an intuitive Streamlit interface for seamless real-time user interactions
LangChainRAGChromaDBLLMsStreamlit

Technical Skills

AI & LLMs

LangGraphLangChainRAGMulti-Agent SystemsPrompt EngineeringFine-tuningLoRA

Cloud & AWS

AWS BedrockAWS SageMakerAWS EKSAWS RDSAzureCloud Architecture

DevOps & MLOps

DockerKubernetesCI/CDGitHub ActionsMLflowHelmGitLab CI/CD

ML & Deep Learning

PyTorchTensorFlowYOLOv11TransformersLSTM / GRUComputer Vision

Inference & Serving

TensorRT-LLMTriton ServerGroqOllamaAWS Bedrock

Languages & Tools

PythonGitStreamlitKafkaSparkElasticsearch

Education

Master of Science in Generative AI

AIvancity School — Paris, France

Oct 2025 — Present

Eduniversal Innovation Award — The New Standard

Software & AI Engineering Degree

INSAT — Tunis, Tunisia

Sep 2020 — Sep 2025

Top 1 Engineering School in Tunisia

Get In Touch

Interested in collaborating or have a question? Feel free to reach out.

Location

Paris, France