Internship – Optimization of a RAG

il y a 5 jours


Palaiseau, Île-de-France Osborne Systems Temps plein

Osborne Systems
is a deep-tech software company building a SaaS platform that automates and standardizes the engineering of industrial flow-measurement systems for critical energy projects.

We help engineering teams reduce design time, errors, and compliance risks across the full project lifecycle.

Within this framework, the company is designing an
agent-based system
dedicated to the extraction and exploitation of technical information from complex, unstructured documents.

This system integrates
advanced parsing techniques (OCR, VLM), hybrid approaches (regex + LLM), business rules, and retrieval-augmented generation (RAG)
to ensure robust, scalable, and high-performance information processing.

Osborne Systems has joined the
École Polytechnique incubator
, providing a demanding and stimulating technical environment.

Missions

The intern will contribute to the
design, optimization, and improvement of the entire architecture
of the agent-based system. The missions notably include:

·     
Optimization of document parsing and structuring strategies
(chunking, metadata extraction, and normalization)

·     
Analysis, comparison, and selection of embedding methods
tailored to industrial technical data

·     
Implementation of query augmentation, rewriting, and retrieval techniques

·     
Design and development of improvements in Python
(LangChain, custom pipelines)

·     
Evaluation of system performance
(result quality, robustness, latency, and scalability)

·     
Contribution to the improvement of all system components
: agents, extraction pipelines, orchestration, and hybrid rule-based/LLM approaches

·     
Proposals for architectural improvements
to enhance reliability, maintainability, and performance

Required / appreciated skills

·     Very strong foundations in
Python

·     Knowledge of
parsing techniques
(OCR, VLM)
and
hybrid systems
(regex + LLM)

·     Knowledge of
LLMs
and
RAG systems

·     Experience with or strong interest in
LangChain

·     Ability to work on complex, production-oriented AI systems

·     Rigor, autonomy, and analytical mindset

Desired profile

·     
Master's
student or
engineering school
student (AI, computer science, data science)

·     Interest in
industrial applications of AI

·     Sensitivity to system performance and reliability issues

Internship location and duration

·     École Polytechnique – Drahi Xnovation Center

·     Duration:
6 months

Application / Contact

Interested candidates are invited to send an email including:

·     A
CV

·     A
short motivation email
describing their background, experience, and interest in generative AI




  • Palaiseau, Île-de-France Osborne Systems Temps plein

    Osborne Systemsis a deep-tech software company building a SaaS platform that automates and standardizes the engineering of industrial flow-measurement systems for critical energy projects.We help engineering teams reduce design time, errors, and compliance risks across the full project lifecycle.Our team has joined the École Polytechnique incubator,...


  • Palaiseau, Île-de-France Pasqal Temps plein

    About PasqalPASQAL designs and develops Quantum Processing Units (QPUs) and associated software tools.Our innovative technology enables us to address use cases that are currently beyond the reach of the most powerful supercomputers; these cases can concern industrial application challenges as well as fundamental science needs.In addition to the exceptional...


  • Palaiseau, Île-de-France Pasqal Temps plein

    About PasqalPASQAL designs and develops Quantum Processing Units (QPUs) and associated software tools.Our innovative technology enables us to address use cases that are currently beyond the reach of the most powerful supercomputers; these cases can concern industrial application challenges as well as fundamental science needs.In addition to the exceptional...


  • Palaiseau, Île-de-France Pasqal Temps plein

    About PasqalPASQAL designs and develops Quantum Processing Units (QPUs) and associated software tools.Our innovative technology enables us to address use cases that are currently beyond the reach of the most powerful supercomputers; these cases can concern industrial application challenges as well as fundamental science needs.In addition to the exceptional...


  • Palaiseau, Île-de-France Pasqal Temps plein

    About PasqalPASQAL designs and develops Quantum Processing Units (QPUs) and associated software tools.Our innovative technology enables us to address use cases that are currently beyond the reach of the most powerful supercomputers; these cases can concern industrial application challenges as well as fundamental science needs.In addition to the exceptional...


  • Palaiseau, Île-de-France CEA Temps plein

    General information Organisation The French Alternative Energies and Atomic Energy Commission (CEA) is a key player in research, development and innovation in four main areas :• defence and security,• nuclear energy (fission and fusion),• technological research for industry,• fundamental research in the physical sciences and life sciences.Drawing...


  • Palaiseau, Île-de-France PASQAL Temps plein

    About PasqalPasqal designs and develops Quantum Processing Units (QPUs) and associated software tools.Our innovative technology enables us to address use cases that are currently beyond the reach of the most powerful supercomputers; these cases can concern industrial application challenges as well as fundamental science needs.In addition to the exceptional...


  • Palaiseau, Île-de-France Pasqal Temps plein

    About PasqalPasqal designs and develops Quantum Processing Units (QPUs) and associated software tools.Our innovative technology enables us to address use cases that are currently beyond the reach of the most powerful supercomputers; these cases can concern industrial application challenges as well as fundamental science needs.In addition to the exceptional...

  • Stage en cryptologie H/F

    il y a 5 jours


    Palaiseau, Île-de-France CEA Temps plein

    Informations générales Entité de rattachement Le CEA est un acteur majeur de la recherche, au service des citoyens, de l'économie et de l'Etat.Il apporte des solutions concrètes à leurs besoins dans quatre domaines principaux : transition énergétique, transition numérique, technologies pour la médecine du futur, défense et sécurité sur un...

  • Public Policy Analyst

    il y a 4 jours


    Palaiseau, Île-de-France Pasqal Temps plein

    About PasqalPASQAL designs and develops Quantum Processing Units (QPUs) and associated software tools.Our innovative technology enables us to address use cases that are currently beyond the reach of the most powerful supercomputers; these cases can concern industrial application challenges as well as fundamental science needs.In addition to the exceptional...