When we think of circuitry, people tend to think of one of two things: Either fairly large discrete components that will balance comfortably on the tip of your finger (image credit: Creatively Maladjusted), or slabs of plastic and ceramic encapsulating integrated circuits which are comprised of millions upon millions of components. At the time I write this article we can fabricate circuitry on a scale of about 14 nanometers and in about two years we'll be able to reliably build circuitry around 10 nanometers in size, which is significantly bigger than the atoms of the elements used in chip manufacture, which are about 0.20 nanometers in size. So, it came as something of a surprise when two separate research teams, one at Argonne National Laboratory and the other at the University of California at Berkeley announced that they had successfully fabricated transistors just three atoms thick. Transistors are comprised of three layers of material, layered so that their electrical properties alternate and function in the same fashion as the humble toggle switch. The research terms have successfully constructed transistors which are as close to two dimensional as is possible right now. This means that they can be made much, much smaller than is common right now and thus can be packed together far more densely. Rather than selectively contaminated silicon the Argonne team used graphene, tungsten diselenide and boron nitride and fabbed the transistors atop a flexible plastic substrate using standard lithographic techniques. The U.Cal Berkeley team built their transistors using molybdenum disulfide instead of tungsten diselenide. Fabrication techniques are still immature, so it's probably going to be a couple of years before we start seeing anything on the market built at this scale. Knowing that multiple compounds are feasible, however, implies that a "best of breed" combination of compounds should emerge, which also means that multiple groups will start trying different combinations and techniques to see what works best under different circumstances.
On a somewhat bigger scale (but still minute as anyone would reckon it) we have cellular microsurgery, or the act of performing surgical techniques on individual cells of larger organisms. As one might expect this is a fairly delicate and difficult body of techniques which involve tools such as excimer lasers and carefully shaped glass micropipettes with tips that are all but invisible to the naked eye. With such tools and a lot of patience you can, in fact, inject and remove DNA from the nuclei of the cells and perfom gene therapy on individual cells. If the cells are zygotes one can, in theory, perform germline engineering upon an organism while it sufficiently immature to be only a few cells in size because it is exceedingly difficult to modify every cell in a fully grown complex organism (like a mouse, or a rabbit, or a human for that matter). This requires no small amount of luck because sometimes the cells pop and sometimes they just stop functioning after they've been operated on. A research team at Bringham Young University has invented a device a little larger than a single cell which is capable of reliably injecting DNA into single cells without killing them using electrostatic principles to exploit the patterns of electrical charges of DNA. Bits of DNA can be injected into the cells without needing to pump them full of fluid which can cause them to rupture; presumably, bits of DNA could also be extracted the same way. The research team used mouse zygotes as its test subjects. The process they implemented has been dubbed metamorphic nanoinjection and involves a microscopic needle (referred to as a lance) about ten micrometers in size. 77.6% of the cells experimented upon survived surgery with the nanoinjector and went on to replicate normally, as opposed to about 54.7% of cells operated on with standard microsurgical techniques and implements. I wish I knew how they built it (the Gizmodo article has a video with a nifty animation of the device operating) but I haven't had time to dig up any more documentation. I think it's safe to say that if this technique was coupled with other methods of infiltrating individual cells we'd have a solid body of techniques to begin building applied synthetic biology on top of.