Search based software engineering’s (SBSE) origins stem from the artificial intelligence (AI) community, where researchers have developed a plethora of optimization algorithms that mimic phenomena in the natural world, something that was recognized by early SBSE adopters.  SBSE has grown from a small community of visonaries who were convinced that hard computational problems in software engineering should be solved heuristically, to a thriving and broad research community. As such, over the past 15-20 years the SBSE community has produced mature test generation tools, spawned the field of APR and GI and embedded itself into almost all of the major software engineering venues. As this has happened the SSBSE venue has evolved as a unique place to talk about these powerful techniques, to diversify our ideas and to converge on new approaches, while at the same time training the next generation of SBSE researchers.  However, there is a new tide rising out of the AI community and we now see machine learning (ML) as its representative face, perhaps leaving SBSE (and maybe SSBSE) behind.  But this does not necessarily mean the end of SSBSE. In fact, ML can leverage SBSE and SBSE ought to incorporate ML.  Hence, I see a future with hybrid ML/SBSE approaches, hyper heuristic (self*) algorithms, and the mixing of human intelligence with automation. In this talk I will reflect on the past of SBSE and propose my vision for a world of ML-enabled-SBSE.