Christian Kästner

I am looking for a postdoc (and also potentially interested undergraduate or visiting students) for a project on improving evolution and configuration of robotics software. We pursue a strategy based on sensitivity analysis, combining machine learning with static or dynamic software analysis, roughly following the line of work of our FSE 2015 paper (Performance-Influence Models for Highly Configurable Systems, with Norbert Siegmund, Alex Grebhahn and Sven Apel), also recently outlined in a workshop paper (Sensitivity Analysis For Building Evolving & Adaptive Robotic Software, with Prasad Kawthekar). The position is part of a large collaborative effort sponsored by DARPA.

Here is the actual call:

Multiple post-doctoral scholar positions on software engineering and 
programming language topics are available in a project on “Intelligent 
Model-Based Adaptation of Robotics Systems” in the School of Computer 
Science at Carnegie Mellon University (CMU) in Pittsburgh, PA. The 
project is a collaboration involving CMU faculty Jonathan Aldrich, David 
Garlan, Christian Kaestner, Claire Le Goues, and Manuela Veloso.

The research project concerns the construction and analysis of robotics 
software with respect to various quality attributes like performance and 
mission success, the automated adaptation and repair of that software, 
and the automated understanding of properties of that software. There 
are many interesting research directions within the project, including 
sensitivity analysis to understand the influence of changes in large 
configuration spaces (machine learning, sampling), the analysis of 
interactions among various changes (static or dynamic analysis, 
instrumentation), the integration of software architecture within the 
language to ease architecture-based adaptation, and the definition of 
domain-specific languages that facilitate adaptation in robotics
systems.

Candidates should have, or shortly expect to receive, a doctoral degree 
in computer science or a related field.  They should have a background 
in either static or dynamic program analysis, programming languages, 
machine learning, or robotics software (ROS). Skills in building 
analysis tools and automating evaluations (e.g. shell scripts) are 
expected. Candidates interested in this position should email a CV 
together with a brief description of research interests and a list of 
two to three references to Christian Kaestner (kaestner@cs.cmu.edu).

Possible postdoc advisors are primarily Jonathan Aldrich or Christian 
Kaestner. The initial contract will be for 12 months. Review of 
applications will begin upon receipt and continue until all positions 
are filled.

            

Please send applications or possible leads by email.