Thesis: Cloud agent-based software development workflows

High Level Description

As software development becomes increasingly agent-based, cloud AI agents can pick up a ticket and carry it through refinement,

planning, implementation, testing, and review. However, to ensure quality and correct implementation developers still need to

check that requirements are understood, quality holds, and design choices fit the project. This thesis aims to design and

implement a reusable cloud-agent based development process with checkpoints where developers can inspect progress, correct

decisions, and approve changes before the work continues.

Project Description

In this thesis, design and implement an automated agentic workflow using cloud-based agents. The workflow should orchestrate

multiple agents to autonomously manage the end-to-end process of transforming a human-written feature request into a completed

implementation. This process includes retrieving a ticket describing the requested feature, refining the requirements, planning

the implementation, integrating the changes into an existing codebase, creating and executing tests, and reviewing the results.

To ensure that the agents correctly interpret the requirements and make appropriate decisions, the workflow should incorporate

human-in-the-loop checkpoints at key stages. These checkpoints should allow a human operator to review, approve, or reject the

agents’ decisions and proposed actions when necessary. The workflow should also provide clear and structured reporting that

enables the operator to understand the changes made and the rationale behind them. This reporting should facilitate switching

between multiple concurrent implementations while minimizing the cognitive overhead associated with context switching.

Since agentic workflows can enable developers to manage multiple projects more efficiently and establish new projects more easily,

the proposed solution should be straightforward to integrate into both existing and new projects. The prototype will be evaluated

through practical testing, focusing on usability, integration, reusability across projects, and the effectiveness of the

human-in-the-loop checkpoints in supporting oversight and control.

Purpose

In this thesis, investigate these questions:

• Design and implement a reusable cloud-agent based process that can develop software from feature request to reviewed

implementation.

• Evaluate usability, integration, reuse, and checkpoint effectiveness in practical use.

An Exciting Journey with Knightec Group

Semcon and Knightec have joined forces as Knightec Group. Together, we are Northern Europe’s leading strategic partner in

product and digital service development. With a unique combination of cross-functional expertise and a holistic business

understanding, we help our clients realize their strategies – from idea to complete solution.

Practical Information

This is a thesis position, located at our office in Sundsvall. Start date January or March 2027.

Please submit your application as soon as possible, but no later than 2026-11-30. If you have any questions, you are welcome to

contact johanna.edstrom@knightec.se. Note that due to GDPR, we only accept applications through our careers page.