Stijn Heymans, PhD
Summary
Currently, I’m a co-founder and developer of Sheldonize, where we aim to automate the scheduling of personal tasks using intelligent heuristics and algorithms.
From 2011 to 2014, I was a computer scientist at the Artificial Intelligence Center of SRI International (the inventors of the mouse in the 60s and Apple’s Siri more recently) working on Question Answering in the context of an intelligent Biology textbook, by means of Knowledge representation and reasoning techniques. Further in the past, my work focused on Semantic Technologies. For more details check out my LinkedIn profile.
Software
|
Sheldonize 2014+ |
This is a web application that is a complete rewrite of todopl. It's a task manager that automatically schedules your to-dos on your calendar. It checks your existing meetings, and plans your work around them, while making sure deadlines for your tasks are met. I did all development. [Django (Python), Bootstrap] |
|
ETB Datalog engine 2013 |
The Evidential Tool Bus (ETB) is a distributed framework for tool integration for the purpose of building and maintaining assurance cases. ETB employs Datalog as a metalanguage both for defining workflows and representing arguments. I implemented the new core Datalog engine for the Evidential Tool Bus; more details on the theory of that engine can be found in the SAS 2014 paper. [Python] |
|
todopl 2012+ |
This is a desktop application that instead of just maintaining lists of tasks also schedules these tasks on a schedule, taking into account meetings you have, deadlines, etc. I did all development. [Clojure, SWI-Prolog] |
|
RAVE/Sunflower 2012-2014 |
RAVE/Sunflower is an IDE for rule- and ontology editing. I wrote the tracer/debugger for RAVE/Sunflower, as well as several UI components in the Query module. [Flora-2, Java] |
|
AURA 2011-2013 |
The Automated User-centered Reasoning and Acquisition System (AURA) allows subject matter experts, for example, biologists, physicists, and chemists, and scientific educators to formulate knowledge and questions to query that knowledge. I contributed to and maintained large chunks of code in the Question Answering module of AURA. [Common Lisp] |
|
mymedintel 2011 |
This was a prototype that slurped in FDA Adverse Event data as well as your Google Health profile and determined whether you where currently at risk of any drug interactions based on your medications. [SWI-Prolog] |
|
oasp-r 2009 |
A Reasoner for Simple Conceptual Programs under the Open Answer Set Semantics. The theory behind this reasoner has been written down in a RR 2009 paper. [B-Prolog] |
Publications
You can find bibliographic details for all of these at Bibsonomy.