RnR TechMed Laboratories exists to shorten the distance between a candidate molecule and a treatment that actually reaches a patient.
Most of drug discovery’s cost and time is lost late — in the wet lab, in animal studies, in clinical trials — finding out that a promising candidate doesn’t behave the way a clean, digital model predicted.
We started RnR TechMed Laboratories to close that gap at the source: by building a hardware platform that emulates the cell’s actual noisy, asynchronous, non-linear behavior, so risk gets identified while a candidate is still just data — not yet a tissue sample.
Reuben is an engineer by training and profession, with a background in high-precision analog circuit design and the design of industrial control systems. He was mentored by the late Dr. James K. Roberge of MIT Lincoln Laboratory while working at MIT from 1999–2004 — work funded by DARPA, the U.S. Defense Advanced Research Projects Agency.
His technical strengths span ultra-high-precision analog VLSI systems, signal-extraction circuitry for adverse signal-to-noise-ratio regimes, and the design of integrated systems with cognitive capabilities — work he sees as a viable alternative to current approaches built on large language models.
Reuben was strongly influenced by Roberge, whose mentorship shaped his early foundational work in spike-driven analog VLSI systems in bipolar semiconductor technology — the lineage this platform’s spiking neural hardware builds on.
Rajeswaran is an engineer who completed his MS at Texas A&M University, where his thesis on real-time systems investigated logical clocks for ordering system events. He went on to work at IBM (Austin) and McAfee in California, and after returning to India, at Cognizant, HCL, and Capgemini. He holds an MBA from IIM Calcutta.
At Capgemini, he led the AI Center of Excellence and was part of the AI Futures Lab. He was an integral part of the “Knights of Knowledge” knowledge-graph community, working closely with Dr. Robert Engles, and built multiple AI-driven solutions across the firm. He is currently pursuing a PhD in large language models in the healthcare domain.
He has published multiple papers and is the creator and maintainer of the R package proportion. Connect with him on LinkedIn.
We’d rather build hardware that respects biological noise than software that pretends it isn’t there.
When the right answer requires new hardware, not a bigger model, we build the hardware.
Explainability isn’t a feature — it’s what lets a researcher trust a result enough to act on it.
Analog circuit designers, computational biologists, and machine learning researchers, working together on one platform.
See open roles