Internship details

• Duration: 5-6 months

• Start date: Flexible, 2026

• Location: Toulouse

• Team: IT

• Internship subject: Assess on-Premise AI Solutions for Code Assistance & Information Retrieval

• Compensation: 1000€ gross, tickets restaurant, CSE

Internship Context

EasyMile is exploring the deployment of on-premise AI solutions to strengthen its engineering and support activities. Under the supervision of the IT Manager, the intern will join a cross-functional team (R&D, IT, Support) to prototype and evaluate how AI models (on the shelves and / or augmented and / or fine-tuned) can improve developer productivity and technical knowledge access, while evaluating the viability of deploying on-premise GPU infrastructure compared to SaaS models, in terms of costs, performance, and data governance.

Internship Goal

The objective of this internship is to evaluate and prototype AI-based solutions for:

• Code assistance: autocompletion, code review, documentation generation.

• Information retrieval: predefined queries across heterogeneous sources (Confluence, JIRA, FluidTopics, internal documentation).

  • Example: “As level 3 support, I want to know all the information relating to XX topic.”

The internship will also explore advanced approaches such as Retrieval-Augmented Generation (RAG) and fine-tuning of models on internal corporate data.

Missions

Under the supervision of the IT Manager, the intern will:

• Benchmark and compare open-source and commercial on-premise AI frameworks (e.g. LLMs, vector databases, orchestration frameworks) balancing costs and performance (latency, throughput, energy efficiency, and scalability).

• Prototype code assistance features integrated with developer workflows (IDE plugins, Git hooks, etc.).

• Design and test search pipelines across multiple internal knowledge sources.

  • This will require working with APIs, building connectors, and handling data retrieval from heterogeneous sources.

• Experiment with RAG and fine-tuning approaches for domain-specific tasks.

• Document findings, trade-offs, and recommendations for production adoption.

• Present regular progress updates to R&D and IT leadership.

There is no typical profile at EasyMile, we all come from different backgrounds and that is what makes us strong! Don’t hesitate to apply if you are motivated and interested by innovative transportation and technologies.

We are looking for:

• Education: Student in computer science, data/AI engineering, or related field.

• Technical skills: Python, knowledge of machine learning/LLMs, experience with APIs and databases; familiarity with NLP or information retrieval is a plus.

• Soft skills: Curiosity, autonomy, ability to synthesize and communicate results clearly.

• Language skills: English and French

• 30 minutes call with the  recruitment team

• Meeting with the team, technical tests

• One hour interview with the tutor and a recruitment team