My name is John Muellerleile, though I also [happily] answer to “jrecursive”, “recursive” or “JR” (as in jay-arr, not “junior”). I’m an engineer and data scientist at Showyou; I am responsible for our data schemas, storage, search and retrieval, recommender engines, scoring algorithms, distributed systems, systems architecture, and — well, hey, it’s a startup right?
In a previous life I was a database engineer at Basho Technologies. My notable R&D work there focused mostly on Riak and Riak Search. I still hack on Riak from time to time, as it’s a versatile platform for several distributed systems configurations that I know extremely well. We also use Riak at Showyou.
I have been some combination of musician, sound engineer and producer for most of my waking life. Other hobbies include taking pictures of the sky, the clouds, and other (hopefully) interesting or somehow moving people, places, things or moments, learning to paint, artificial intelligence, collective behaviors, reconfigurable computing, … — well, math, basically, though I learn more by falling down than instruction, and more often than not I call things what I see them to be rather than whatever they might be called. Sorry about that. I’m working to close that gap.
Here is a little slice of different things, work and otherwise, I’ve done over the years:
* I’ve built a new search & retrieval oriented database on top of Riak (riak_kv) called “Mecha”; you can check out my Mecha presentation slides and a video of my tech talk on Mecha at Boundary. Here are some screen shots of monitoring, browsing data and other functionality; I do have a few explanations of features: system tap, bottom-up query graph language example, integration query language example, query language macro use example, step-by-step example of query graph language, ad-hoc code-injection into the query graph at runtime, sort-merge equijoin example, geospatial query example, utility and system/storage metadata querying, ‘drop bucket’, ‘derive-schema’ for mixed-schema buckets, language support, management/runtime system data
* protograph is a little in-memory “database” for graphs (as in networks), modeled a bit after redis
* I researched, designed and implemented an ANSI SQL compliant query engine, stream processor and supporting systems on top of HBase; the IP is tied up in receivership hell — relationship status: it’s complicated
* My Riak Search presentation at Erlang Factory 2010 San Francisco
* Graphs relating information about “discovered” people extracted from free text
* Discovery & automated research tools
You can find me on twitter as @jrecursive.