Master Thesis: Cost-Effective Document Generation with Code Insights Using Coding Agents
Om jobbet
Join our Team
About this opportunity:
With the rapid advancement in Generative AI-especially ever-evolving LLMs and coding agents-large parts of software development can now be done with far less effort and deep domain knowledge. Coding agents can be used for other tasks apart from code generation, such as generating code documentation, specifications, and even higher-level insights from a codebase. However, "vibe coding" (prompt-and-hope workflows) often breaks down because requirements remain fuzzy in the prompt, so agents guess, drift, and produce inconsistent output. Spec Driven Development (SDD) solves this by putting all requirements into a single, structured spec that every agent follows. That makes generation repeatable, reviewable, and auditable: change the spec, regenerate the code or docs, and everything stays aligned.
This thesis explores how to apply SDD to make AI-produced code documentation and insights reliable enough for real software teams.
What you will do:
- Perform a literature and tooling survey on spec-driven development, coding agent tools, and AI-assisted documentation.
- Analyze how existing SDD kits (e.g. SpecKit/OpenSpec-style approaches) can be adapted to produce documentation and insights automatically from specs and code.
- Design a spec format (or extend an existing one) that links business intent → technical spec → code → generated documentation. The spec is designed based on personas: such as Architect, API Developer/Integrator, DevOps, Onboarding, Release/Project Manager etc.
- Build a prototype pipeline that:Provide and ingests a spec, guides AI/coding agents to follow the specs without hallucination from LLM and generates coherent documentation (e.g. module descriptions, API references, change rationales).
- Collect representative examples from real or sample codebases to test the pipeline.
- Run expert or stakeholder evaluations on the clarity, completeness, and maintainability of the generated documentation vs. baseline/manual docs. Use also LLM-as-judge for the generated output.
- Quantify the potential business/engineering impact (faster onboarding, better reuse, lower knowledge-loss risk) of SDD-based documentation.
- Deliver a practical decision framework for when and how to adopt SDD for documentation and insights.
The skills you bring:
- Currently pursuing a master's degree in Computer science/AI/ML or equivalent field.
- Strong analytical skills and experience in SW Development process.
- Familiarity with Generative AI concepts and tooling and their impact on business value and operations.
- Interest in emerging technologies and their strategic implications for telecom.
- Suitable for 1 or 2 students.
Why join Ericsson?
At Ericsson, you'll have an outstanding opportunity. The chance to use your skills and imagination to push the boundaries of what's possible. To build solutions never seen before to some of the world's toughest problems. You'll be challenged, but you won't be alone. You'll be joining a team of diverse innovators, all driven to go beyond the status quo to craft what comes next.
What happens once you apply?
Click Here to find all you need to know about what our typical hiring process looks like.
Encouraging a diverse and inclusive organization is core to our values at Ericsson, that's why we champion it in everything we do. We truly believe that by collaborating with people with different experiences we drive innovation, which is essential for our future growth. We encourage people from all backgrounds to apply and realize their full potential as part of our Ericsson team. Ericsson is proud to be an Equal Opportunity Employer. learn more.
Primary country and city:
Sweden (SE) || Stockholm
Req ID:
775788
About this opportunity:
With the rapid advancement in Generative AI-especially ever-evolving LLMs and coding agents-large parts of software development can now be done with far less effort and deep domain knowledge. Coding agents can be used for other tasks apart from code generation, such as generating code documentation, specifications, and even higher-level insights from a codebase. However, "vibe coding" (prompt-and-hope workflows) often breaks down because requirements remain fuzzy in the prompt, so agents guess, drift, and produce inconsistent output. Spec Driven Development (SDD) solves this by putting all requirements into a single, structured spec that every agent follows. That makes generation repeatable, reviewable, and auditable: change the spec, regenerate the code or docs, and everything stays aligned.
This thesis explores how to apply SDD to make AI-produced code documentation and insights reliable enough for real software teams.
What you will do:
- Perform a literature and tooling survey on spec-driven development, coding agent tools, and AI-assisted documentation.
- Analyze how existing SDD kits (e.g. SpecKit/OpenSpec-style approaches) can be adapted to produce documentation and insights automatically from specs and code.
- Design a spec format (or extend an existing one) that links business intent → technical spec → code → generated documentation. The spec is designed based on personas: such as Architect, API Developer/Integrator, DevOps, Onboarding, Release/Project Manager etc.
- Build a prototype pipeline that:Provide and ingests a spec, guides AI/coding agents to follow the specs without hallucination from LLM and generates coherent documentation (e.g. module descriptions, API references, change rationales).
- Collect representative examples from real or sample codebases to test the pipeline.
- Run expert or stakeholder evaluations on the clarity, completeness, and maintainability of the generated documentation vs. baseline/manual docs. Use also LLM-as-judge for the generated output.
- Quantify the potential business/engineering impact (faster onboarding, better reuse, lower knowledge-loss risk) of SDD-based documentation.
- Deliver a practical decision framework for when and how to adopt SDD for documentation and insights.
The skills you bring:
- Currently pursuing a master's degree in Computer science/AI/ML or equivalent field.
- Strong analytical skills and experience in SW Development process.
- Familiarity with Generative AI concepts and tooling and their impact on business value and operations.
- Interest in emerging technologies and their strategic implications for telecom.
- Suitable for 1 or 2 students.
Why join Ericsson?
At Ericsson, you'll have an outstanding opportunity. The chance to use your skills and imagination to push the boundaries of what's possible. To build solutions never seen before to some of the world's toughest problems. You'll be challenged, but you won't be alone. You'll be joining a team of diverse innovators, all driven to go beyond the status quo to craft what comes next.
What happens once you apply?
Click Here to find all you need to know about what our typical hiring process looks like.
Encouraging a diverse and inclusive organization is core to our values at Ericsson, that's why we champion it in everything we do. We truly believe that by collaborating with people with different experiences we drive innovation, which is essential for our future growth. We encourage people from all backgrounds to apply and realize their full potential as part of our Ericsson team. Ericsson is proud to be an Equal Opportunity Employer. learn more.
Primary country and city:
Sweden (SE) || Stockholm
Req ID:
775788
Ericsson AB
FöretagEricsson AB
Visa alla jobb för Ericsson AB