I have designed & implemented a new engine for Hadoop MapReduce which is extremely fast and fault-tolerant. I have done a lot of experiments using Hadoop to compare its performance with my implementation. I also enjoy writing bash and python scripts to automatize experiments and analyze logs.