Master Thesis: Spec-to-Code AI Development
publicerat av Ericsson
- Företag
- Ericsson
- Ort
- Karlskrona
- Sista ansökningsdag
- Ej angiven
- Nivå
- Master/civilingenjör (30 hp)
- Antal studenter
- 1–2 studenter
- Period
- 30 ECTS credits per student; start by agreement.
- Passar program
- Computer Science, Software Engineering, Computer Engineering
Beskrivning
This thesis explores whether machine-readable software specifications, AI assistance and automated validation can produce production-ready code while improving development speed and quality. Based at Ericsson Fintech in Karlskrona, the project focuses on the specification-centric development workflow for Ericsson Offline Payments. It examines how this workflow performs in practice and how it could be improved.
The assignment
- Identify where effort and lead time build up during specification writing, code generation, manual integration, review and validation.
- Evaluate specification-driven, AI-assisted development against a code-first workflow using comparable tasks.
- Test potential improvements to generator coverage, AI guardrails, validation gates or traceability.
- Define and apply metrics for lead time, rework, consistency and defects.
- Build and demonstrate selected improvements in a realistic software development environment.
- Share findings and recommendations with an Ericsson development team.
Who we are looking for
- Master’s student in Computer Science, Software Engineering, Computer Engineering or a related discipline.
- Strong foundations in programming and software engineering, with confidence using Git and modern development workflows.
- Interest in generative AI, developer tools, requirements engineering or software development automation.
- An experimental approach and enthusiasm for building, measuring and comparing solutions.
- Experience with REST/OpenAPI, PlantUML, code generation, CI/CD or AI coding assistants is an advantage.
What we offer
- Hands-on experience with generative AI, specifications and automation in a production-oriented Fintech development environment.
- Collaboration with experienced software developers and the opportunity to refine the thesis focus together with supervisors.
- Scope to develop ideas and prototypes that may influence future working methods in large-scale Fintech R&D.
Practical details
Location: Karlskrona. Open to 1–2 students, with 30 ECTS credits per student. Start date to be agreed.
Förkunskapskrav
Master-level studies in Computer Science, Software Engineering, Computer Engineering or a related field; solid programming and software engineering fundamentals; familiarity with Git and modern development workflows; and an interest in AI-assisted or automated software development. Experience with REST/OpenAPI, PlantUML, code generation, CI/CD or AI coding assistants is beneficial but not required.
Texten är en sammanfattning av Ericssons annons. Läs hela annonsen och ansök hos Ericsson