Portrait of Mehdi Acheli

Ph.D. · Postdoctoral Researcher · CEA

Mehdi Acheli: Intelligence across
the continuum.

I’m Mehdi Acheli, an AI researcher following intelligence from event logs to the edge and IoT, where data is rich and resources are scarce.

$ 

01 About

I am an AI researcher with a PhD in computer science, driven by research and innovation in data science.

My doctoral work at Université Paris-Dauphine-PSL introduced me to process science. I developed methods to discover behavioral patterns in the event logs of flexible processes, where traditional process models fail to capture how work actually unfolds. I then moved to industry R&D, first predicting process behavior at Logpickr, then at Sogeti, part of Capgemini, where I led a research project on fraud detection: identifying anomalies in financial processes from historical event-based data.

Returning to academia as a postdoctoral researcher at Télécom SudParis, where I supervised PhD students and interns, I turned to the Cloud-Edge-IoT continuum. There, too, processes are central: a service composition is itself a process, coordinating services with dependencies and execution order. My work followed this process through its lifecycle across cloud, edge and IoT devices: generating compositions from user intent with large language models, monitoring them for anomalies as they run, and using language models again to renegotiate service-level agreements when anomalies occur. Monitoring the continuum also means running learning models close to the devices themselves, which led me to work on training models on resource-constrained devices at its lower layers. Beyond the continuum, I worked with a cybersecurity firm on predicting security ratings from companies’ rating histories, another form of event-based historical data, and drew on my experience with language models to automate third-party risk assessment questionnaires.

Today, as a postdoctoral researcher at CEA, I focus on adapting learning models to the edge and IoT. This line of work, which began at Télécom SudParis with lightweight anomaly detection through knowledge distillation and a dual-model training approach, now continues with my research on federated learning across devices with heterogeneous resources.

02 Research interests

Federated & edge learning

Training and optimisation of models in resource-constrained environments (Edge, IoT, embedded); federated learning under resource heterogeneity; knowledge distillation.

Cloud-Edge-IoT continuum

Anomaly detection, orchestration process, its monitoring and adjustment, and agentic AI for the Cloud-Edge-IoT continuum.

Process science

Mining event logs, behavioural pattern extraction, causal inference, and next-activity prediction inspired by NLP and computer vision.

Graph learning & AutoML

Generative graph diffusion for anomaly detection, graph neural networks for fraud detection and predictive process monitoring, and reinforcement-learning-based neural architecture search (RL-GraphNAS).

Security rating

Security rating systems and supplier risk assessment.

03 Projects

PEPR Cloud2023 – 2029

SPIREC

In the highly heterogeneous Cloud-Edge-IoT continuum, monitoring, anomaly detection and resource-use prediction are essential to secure applications and guarantee quality of service. SPIREC develops new methods to address these challenges.

Funded by the French Government under the France 2030 national investment plan.

InriaIMTCNRSCEAUniversité Versailles Saint-Quentin-en-YvelinesUniversité de Lorraine

My role

  • Day-to-day coordination of the work package and scientific point of contact
  • Taking part in the supervision of the PhD students recruited on the project
  • Project Communication Focal Point
  • Research and scientific output (papers, deliverables, etc.)

Objectives

  1. 01Data collection and modeling for supervision and anomaly detection
  2. 02Monitoring for large infrastructures and applications in heterogeneous IEC continuum environments
  3. 03Distributed multi-criteria artificial intelligence for resource use prediction
  4. 04Applications, validation, implementation and integration

PEPR Cloud2023 – 2030

TARANIS

In the Cloud-Edge-IoT continuum, heterogeneous, distributed and energy-constrained resources make cloud management far more complex. TARANIS abstracts applications and resources to automate their management and optimize resource use.

Funded by the French Government under the France 2030 national investment plan.

InriaCNRSIMTUGACEAUniversité de RennesENS LyonUniversité Claude Bernard Lyon 1Université de LilleINSA Rennes

My role

  • Occasional representative of the team, on administrative and scientific matters
  • Occasional supervision of a PhD student recruited on the project
  • Research and scientific output (papers, deliverables, etc.)

Objectives

  1. 01Investigate the complexity of Cloud-Edge application and infrastructure models
  2. 02Investigate deployment and reconfiguration issues
  3. 03Extend orchestrators for the Cloud-Edge-IoT continuum
  4. 04Revisit optimization issues associated with the use of Cloud-Edge-IoT infrastructures

Bpifrance2024 – 2026

EVA

EVA aims to deliver continuous security assessment that is simple, automated and affordable, encouraging organizations of every size to strengthen their security posture. It builds on the partner company’s existing assessment solutions and adds new ones.

Funded by Bpifrance.

Cybersecurity company (anonymized)IMT Télécom SudParis

My role

  • Taking part in the supervision of the PhD student recruited on the project
  • Research and scientific output (papers, deliverables, etc.)
  • Contributing to the administrative management of the project

Objectives

  1. 01Build a mechanism that suggests answers to security questionnaires (vendor assessments, DORA, NIS2, brokers…) from a small dataset
  2. 02Improve the mapping of a company’s assets
  3. 03Uncover the exposure of companies and their executives to cyber risks caused by data leaks
  4. 04Refine the security rating algorithm so that ratings stay relevant and evolve despite constant change

04 Experience

  1. Oct 2026 – Present

    Postdoctoral Researcher — Federated Learning

    CEA

    Federated learning under resource heterogeneity.

  2. Sep 2024 – Sep 2026

    Postdoctoral Researcher — AI

    Télécom SudParis

    • Project management; coordinated a work package on PEPR Cloud SPIREC. Contributor to PEPR Cloud Taranis and a Bpifrance project (research and management).
    • Supervised Ph.D. students: anomaly detection and SLA management in the Cloud-Edge-IoT continuum (knowledge distillation, agentic AI), security rating systems, and supplier risk assessment.
    • Supervised an intern on anomaly detection in dynamic graphs using generative graph diffusion as an adaptation of Stable Video Diffusion.
    • Individual research on a new technique for optimising model training in resource-constrained environments (Edge, IoT, embedded).
  3. Sep 2022 – Aug 2024

    R&D Project Manager

    Sogeti, part of Capgemini

    • Led data science R&D projects using agile methods; mentored consultants, apprentices, and interns.
    • Literature review, research, and scientific communication.
    • Built ML, DL, and RL solutions: a meta-model generating neural networks via RL-GraphNAS, graph-neural-network anti-fraud, data imbalance, synthetic data generation.
  4. Dec 2021 – May 2022

    R&D Engineer

    Logpickr, iGraphx

    Predicting consequent activities in a process, drawing on NLP and computer vision.

05 Publications

Highlighted namesPh.D. students whose papers were written under my co-supervision PreprintsHidden by default; use the Preprints button to include them CORE AConference rank from the CORE portal (ICORE2026); journal quartile from Scimago

06 Teaching

About 230 hours of lab and tutorial teaching (TD/TP) at Université Paris-Dauphine-PSL during and after my Ph.D.

YearCourse UnitTypeVolume
2021–2022Bases de Données RelationnellesTD/TP52,5 h
2020–2021Systèmes d’ExploitationTD/TP40,5 h
C++TD/TP21,94 h
2019–2020Outils en InformatiqueTP24 h
Bases de Données RelationnellesTD/TP16,13 h
Programmation CTD/TP10,5 h
2017–2018Outils en InformatiqueTP12 h
Bases de Données RelationnellesTD/TP52,5 h

07 Supervision

Ph.D. students · co-supervision

SPIREC

Farah Abid

Data science for resource monitoring, SLA renegotiation and anomaly detection in the Cloud-Edge-IoT continuum

SPIREC

Abderrahmane Boucenna

Anomaly detection and monitoring in the Cloud-Edge-IoT continuum: from data models to intelligent detection and optimal workload placement

EVA

Nada Hanad

Contextual rating system for the automated assessment of IT infrastructures, integrating AI models

TARANIS

Wided Meflah

Automated orchestration across the Cloud-Edge-IoT continuum

Master’s students

M2 work-study2022 – 2023

Neila Khermane

Master II Vision et Machine Intelligente, Université Paris Cité

Exploring neural networks with modular architecture in the context of bank fraud detection

M2 work-study2022 – 2023

Fatima Zahra Remmach

Master 2 Machine Learning for Data Science, Université Paris Cité

Financial fraud detection with Graph Neural Network methods

M2 internship2024 – 2025

Sofien Mohamed

Master Professionnel en Informatique, Horizon School of Digital Technologies

Anomaly detection in dynamic graphs using generative graph diffusion

08 Academic service

Organization

Reviewer · Conferences

Reviewer · Journals

09 Education

  1. 2021

    Ph.D. in Computer Science

    Université Paris-Dauphine–PSL

    Data mining on business-process logs; behavioural pattern extraction and feature engineering; data analytics, correlation studies and causal inference. Taught C, C++, Unix, RDBMS, HTML, MS Access.

    Thesis: Behavioral Pattern Mining for Flexible Processes, supervised by Pr. Daniela Grigori. Read thesis ↗

  2. 2017

    State Engineer in Computer Science, specialised in Computer Systems and Machine Learning

    Higher National School of Computer Science | LIAS (ENSMA)

    Graduation project: benchmarking of databases (NoSQL, distributed RDBMS, Hive) and development of data and query generators. Public outreach through tutorials.

Certifications

Professional Scrum Master I (PSM I) Scrum.org · Verify certificate ↗ (enter contact address)
TOEIC Score: 990/990

10 Skills

Machine learning
Random forest, SVM, clustering, CNN, RNN, GNN, Transformers, auto-encoders, reinforcement learning, NLP, LLMs, computer vision, generative diffusion
Development
Python (PyTorch, LangChain, NumPy, pandas, scikit-learn, Gym, CUDA), Java, C, C++, C#, VHDL, SQL, R, LaTeX, BPMN
Infrastructure
Docker, Vagrant, VMware, VirtualBox, AWS; GitLab, GitHub, SVN
Databases
MySQL, MongoDB, Oracle, PostgreSQL, Postgres-XL, Citus
Management
Scrum (PSM I certified), Kanban
Languages
Arabic (native), Kabyle (native), French (bilingual), English (bilingual, TOEIC 990/990)

11 Contact

Interested in collaborating, supervising, or reviewing? Let’s talk.

mehdi.acheli@gmail.com