Description du poste

Internship

Field: AI & Software Engineering
Duration: 6 months
Paid internship with a flexible workplace model allowing remote and on-site work where possible.

Project objectives and responsibilities

Study the current incident lifecycle from bank user to support and ticketing tool, then design the target automated flow.
Develop an in-app “Report an incident” feature in ACP that captures screenshots, user action trails, error logs and stack traces, module, version, environment, and related application ID.
Build an LLM-based engine that converts the user’s description and captured context into a structured ticket containing the title, reproduction steps, expected and actual behavior, suggested severity, priority, and module classification.
Detect duplicate incidents through semantic similarity against existing tickets using embeddings and a vector database, and suggest known workarounds.
Integrate with a ticketing tool through a REST API, such as Jira or ServiceNow, to create tickets automatically and return their status to the user.
Ensure banking-grade compliance through personal-data masking, confidentiality, and an audit trail.
Measure impact, including ticket completeness and triage time, then document and demonstrate the solution.

Final-year engineering or master’s student in Software Engineering, AI, or Data Science.
Good skills in Java and a web front-end framework such as Angular or React, plus Python.
Knowledge of NLP and LLMs, including RAG, embeddings, and prompt engineering.
Experience integrating REST APIs; familiarity with Jira or other ticketing tools is a plus.
Awareness of data privacy and security concerns.
Analytical mindset, rigorous, and comfortable in French and English.

Description du poste

Field: AI & Software Engineering
Duration: 6 months
Paid internship with a flexible workplace model allowing remote and on-site work where possible.

Study the current incident lifecycle from bank user to support and ticketing tool, then design the target automated flow.
Develop an in-app “Report an incident” feature in ACP that captures screenshots, user action trails, error logs and stack traces, module, version, environment, and related application ID.
Build an LLM-based engine that converts the user’s description and captured context into a structured ticket containing the title, reproduction steps, expected and actual behavior, suggested severity, priority, and module classification.
Detect duplicate incidents through semantic similarity against existing tickets using embeddings and a vector database, and suggest known workarounds.
Integrate with a ticketing tool through a REST API, such as Jira or ServiceNow, to create tickets automatically and return their status to the user.
Ensure banking-grade compliance through personal-data masking, confidentiality, and an audit trail.
Measure impact, including ticket completeness and triage time, then document and demonstrate the solution.

Final-year engineering or master’s student in Software Engineering, AI, or Data Science.
Good skills in Java and a web front-end framework such as Angular or React, plus Python.
Knowledge of NLP and LLMs, including RAG, embeddings, and prompt engineering.
Experience integrating REST APIs; familiarity with Jira or other ticketing tools is a plus.
Awareness of data privacy and security concerns.
Analytical mindset, rigorous, and comfortable in French and English.