OctoML posted an article about using TVM to compile and optimize a deep learning model to run on an embedded system. Since I am a huge fan of TVM (and wrote some clojure bindings at one point), I want to reshare here in case anyone is interesting in some really cutting edge stuff. https://medium.com/octoml/bringing-tiny-ml-to-the-cloud-connected-iot-with-tvm-2ab800b065b7