Master Thesis - Integrating Agentic AI Model Into Development Workflows

Req ID:499944 

At Alstom, we understand transport networks and what moves people. From high-speed trains, metros, monorails, and trams, to turnkey systems, services, infrastructure, signalling and digital mobility, we offer our diverse customers the broadest portfolio in the industry. Every day, more than 80 000 colleagues lead the way to greener and smarter mobility worldwide, connecting cities as we reduce carbon and replace cars.

Master Thesis - Integrating Agentic AI Model Into Development Workflows

Agentic AI model is becoming a new industrial standard to assist developers in their tasks. The MCP (Model Context Protocol) framework enables the integration of multiple AI agent tools, each with a specific responsibility.  Alstom development process uses different tools to follow-up work, to manage documentation and pieces of source-code.

We are looking for a Master Thesis student to join our team in Västerås!

Currently, Alstom uses different tools for different tasks. These tools operate in silos, making it challenging for developers to access planning data, requirements, and documentation efficiently within their development environment. At the same time, AI assisted coding tools like GitHub Copilot have become popular for improving developer productivity.

Use-case
Alstom has different projects in maintenance, which use the IBM Engineering Jazz toolchain (EWM for planning, DOORS for requirements) and GitLab for source code management. These well-known projects are a good field to build-up an autonomous system able to learn the projects context and then assist the developer in the daily routine work.

The proposed solution could use the developer’s IDE (VS Code) to allow the connection between GitHub Copilot and MCP-based agent tools.

This would allow developers to:
- Retrieve and manage planning items from EWM.
- Access requirements from DOORS.
- Query or download documentation from Serena Dimensions.
- Link changes to GitLab Merge Requests

The student will design and implement a proof-of-concept demonstrating these capabilities and evaluate different agentic workflows to identify the most effective approach.

The objectives would be
- Define and integrate the agent tools in the AI environment using the MCP framework.
- Evaluate different agentic workflows to propose the agent combinations delivering the most accurate code changes.

Topic proposed: Integrating Agentic AI Model Into Development Workflows
Duration: 20 weeks (Master) on site

Background: You are a student for a master software engineering program which have an iterrest in usecases for AI large language models. You are profficient in programming languages such as python, java script, C and C++. Development using github co-pilot agent.

You don't need to be a train enthusiast to thrive with us. We guarantee that when you steo onto one of our trains with your friends and family, you'll be proud. If you're up for the challange, we'd love to hear from you!

Important to note
As a global business, we're an eual-opertunity employer that celebrates diversity across the 63 countries we operate in. We're comitted to creating an inclusive workplace for everyone.