Lehrstuhl für Angewandte Softwaretechnik
Chair for Applied Software Engineering
Sebastian Klepper

Sebastian Klepper

Office: 01.07.42

E-Mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Phone: +49 (0)89 289 - 18213

Technische Universität München
Institut für Informatik I1
85748 Garching bei München
Germany

Research Topics

I’m primarily concerned with continuous and empirical software engineering in highly complex problem domains.

Rationale: Even though it anticipates unknowns and change, typical agile development often doesn’t cut it for extremely complex and unpredictable problems.
In such projects it’s indispensable to work with problems instead of requirements, make data-driven decisions and test hypotheses and results in production.

Teams therefore must integrate an empirical research process into their development workflow. This includes analyzing and breaking down problems, formulating hypotheses and testing them,
designing experiments and running them in production if necessary, evaluating results and making decisions.

This requires a perfect handle on continuous software engineering plus advanced workflows, techniques, architectural patterns, infrastructure, etc.
Even though lots of building blocks and trends exist, compared to Scrum & Co. there’s a lack of formalized and proven methodologies for teams for learn and tailor.

Primary fields of research:

  • Hypothesis-driven development
  • Opportunistic release management
  • Continuous experimentation
  • Continuous software evolution
  • Evolutionary architectures

Underlying techniques and topics:

  • Agile and lean development
  • Continuous software engineering
  • Continuous delivery
  • Release management

Office Hours

Available for in-person meetings on Thursdays. Please contact me via email in advance.

Theses

Open

Master’s Thesis / Guided Research
Empirical Development of Complex Systems
Advisor
Sebastian Klepper
Software that solves complex problems can’t be planned in advance because the solution (and often the problem itself) is largely unknown. This necessitates an empirical approach to continuous software engineering.
Using an evolutionary architecture and hypothesis-driven development, we will gradually build up a system that analyzes and generates textual content, e.g. using machine learning.
Prerequisites: Strong software engineering background and basic familiarity with continuous delivery, distributed systems, and machine learning (esp. natural language processing).

In Progress

Finished

Bachelor’s Thesis
Context-Aware Process Transformation of Anti-Patterns in Agile Software Projects
Advisor
Sebastian Klepper
Author
Özge Soydemir
Date
15.07.2016

Publications Sebastian Klepper's citations