29 Smart Prioritization and Classification PFE

Project overview

• REF : PRX-2026-14 — Build an AI agent to automate and optimize customer journey handling by prioritizing and classifying incoming tasks using LLMs.
• The system must produce concise customer-context summaries and actionable recommendations, proactively detect anomalies in journeys, and trigger system functions (e.g., sendReminderEmail(clientId)) or surface justified alerts to users.

Objectives & responsibilities

• Implement Smart Prioritization and Classification: use large language models to automatically classify incoming tasks by priority and complexity so agents focus on the most critical cases.
• Develop Summary and Action Recommendations: generate concise summaries of customer context and suggest next actions based on history and journey steps.
• Implement Proactive Detection and Alerts (Actionable AI): monitor customer journeys, detect anomalies (such as a client stuck at a stage), and trigger programmatic actions or alerts with justification.

Technical scope & deliverables

• Integrate AI capabilities within a Spring Boot backend using Spring AI and Langchain4J to orchestrate LLM calls and chain-of-thought logic.
• Deliver a working prototype, unit/integration tests, documentation of models and decision rules, and demonstration scripts for automated actions (e.g., sendReminderEmail(clientId)).

Required profile & skills

• Engineer profile; the offer targets 1 Trainee (intern/Trainee) with interest in AI/ML and backend development.
• Desired skills: familiarity with machine learning / NLP concepts, experience or strong interest in Spring Boot, ability to work with LLMs and tools like Langchain4J, and basic software engineering practices (APIs, testing, CI).

Technologies & tools

• Core stack: Spring Boot, Spring AI, Langchain4J.
• Expected additional tools: REST APIs, monitoring/alerting libraries, logging, and standard development tooling (Git, build system).

How to apply

• Apply via the Trainees Platform: https://trainees-platform.proxym-group.net
• In your application, reference PRX-2026-14 and highlight relevant projects or coursework on LLMs, Spring Boot, and backend automation.