Kevin Brew

In 2024, I was promoted to senior hardware engineer with a focus on R&D for materials in AI, logic, and chiplet. My work is focused on the enablement of unit process research, namely focus on contact resistance, thermal conductivity and dielectric constant measurements with short-loop vehicles. Part of my current role leading unit process for BSPDN backside Rho-R&D as part of the IBM-Samsung JDA.

Prior to my work on BSPDN backside contact, I was the lead unit process engineer for AI Hardware and Integration research at the IBM AI Hardware Center in Albany, NY. My work focused on engineering phase-change memory (PCM) and resistive memory (RRAM) materials for inference and training. The materials and devices I've developed, with the collaboration of the world-wide IBM Research team, were incorporated into IBM's first 300 mm multi-level PCM built on 14 nm CMOS for AI compute.

I am also an IBM Master Inventor, inaugurated in 2020 with renewal in 2024, and I lead the Albany Technical Vitality Council; a coalition of researchers focused on increasing the technical eminence of our IBM team.

From 2018 to 2021, I led the ULVAC-IBM joint-development program in Albany and am currently a PI for the SUNY Polytechnic-IBM joint research project to assess new PCM materials.

My prior work starting in May 2017 in Albany was concentrated on Middle-of-the-Line (MOL) material development for next generation silicide contacts and metal interconnects. I received my B.S. in Chemical Engineering in 2010 at the University of Delaware and my Ph.D. in Chemical Engineering at Purdue University in 2015. My doctoral thesis entitled, 'Processing of Thin Film Photovoltaics from Chalcogenide Nanoparticles,' was supervised by Dr. Rakesh Agrawal. As an NSF IGERT Fellow & Associate, I developed lab-scale methods to improve efficincies of CZTSSe solar cells processed on flexible glasses, developed novel ligand-exchange chemistries to reduce carbon contamination in solar absorber films, and used my PV expertise to improve the capacity of lithium ion batteries with nanoparticle selenium. From 2015-2016, I began as a post-doctoral researcher in IBM Watson's neuromorphic computing group where I invented and built a completely new two terminal, solid state electrochemical memsitor (ECRAM) that was able to achieve the holy-grail of near-symmetric resistance switching across a continuum of resistance states. During my post-doc, my team was one of the top winning teams of IBM's cognitive build for the 'What Did I Eat?' cognitive app. My postdoctoral experience and inception to the Albany Semiconductor Research Groups inspired my continuing work on integratable memristive materials for AI computing.