Steven Rudich (born October 4, 1961) is a professor in the Carnegie Mellon School of Computer Science. In 1994, he and Alexander Razborov proved that a large class of combinatorial arguments, dubbed natural proofs were unlikely to answer many of the important problems in computational complexity theory. For this work, they were awarded the Gödel prize in 2007. He also co-authored a paper demonstrating that all currently known NP-complete problems remain NP-complete even under AC0 or NC0 reductions.
Amongst Carnegie Mellon students, he is best known as the teacher of the class "Great Theoretical Ideas in Computer Science" (formerly named "How to Think Like a Computer Scientist"), often considered one of the most difficult classes in the undergraduate computer science curriculum. He is an editor of the Journal of Cryptology, as well as an accomplished magician. His Erdős number is 2.
Andrew's Leap 
Rudich (and Merrick Furst, now a Distinguished Professor at Georgia Tech) began the Andrew's Leap summer enrichment program for high school (and occasionally, middle school) students in 1991. The summer enrichment program focuses mainly on theoretical aspects of Computer Science in the morning, followed by lunch recess, and then an elective -- Robotics, Programming, or Math Theory. Most days, there is also an afternoon lecture by a Carnegie Mellon University faculty member.
To enroll in Andrew's Leap, one must take a specialized test known as The Interesting Test. This assessment is supposed to gauge ability to think outside the box, and aptitude for computer-related math.
- "ACM-SIGACT Awards and Prizes: 2007 Gödel Prize".
- "EATCS: Gödel Prize - 2007".
- Agrawal, M.; Allender, E.; Rudich, Steven (1998). "Reductions in Circuit Complexity: An Isomorphism Theorem and a Gap Theorem". Journal of Computer and System Sciences (Boston, MA: Academic Press) 57 (2): 127–143. doi:10.1006/jcss.1998.1583. ISSN 1090-2724.