I was involved in the creation of the IPI PAN Corpus of Modern Polish, mostly experimenting with the possibilities that the TEI and the CES offered for its architecture. I was also hired as XML architect for the Polish National Corpus.
My current projects include the open-source FreeDict (TEI-encoded translating dictionary repository) and OCTC (TEI-encoded multi-instance stand-off multilingual and partly parallel text corpus <note>I could go on with these attributes, you know -- Syd says I'm articulate ;-)</note>). I feel honoured to have been invited to help with sharing resources between FreeDict and the Apertium project (doing rule-based machine translation). I'm also involved in a Polish Ministry of Science-funded project creating a new Swahili-Polish-Swahili dictionary (N104 050437, directed by Beata Wójtowicz).
Some of my experiences with applying the TEI to large-scale language resources are reflected in my recent Balisage-2010 paper, on "Why TEI stand-off annotation doesn't quite work, and why you might want to use it nevertheless".
With Andreas Witt and Eleonora Litta, I am co-convener of the brand-new TEI Special Interest Group, "TEI for Linguists". I am also running for the TEI Council for the 2011-2013 term (see my candidate statement).