In 2017 I joined the BICEP/Keck program, which maps the microwave sky at degree-scale resolution in search of evidence for cosmic inflation. Motivated by several long-standing problems in cosmology, the theory predicts specific initial conditions for the hot early universe: a nearly scale-invariant, Gaussian spectrum of scalar density perturbations, corroborated in detail by galaxy surveys and observations of the Cosmic Microwave Background (CMB). There is one decisive prediction of inflation that has so far eluded observational confirmation: the presence of primordial gravitational waves seeded by tensor fluctuations, traceable today as B-mode polarization of the CMB. The intrinsic faintness of the signal, bright Galactic emission at the same observing frequencies, and distortion from gravitational lensing, all contribute to making this measurement extremely difficult. After several generations of instruments in the field, the BICEP/Keck program continues to produce world-leading constraints on primordial gravitational waves and is currently deploying the "Stage-3” BICEP Array telescope.

Targeting faint polarization patterns at the nanokelvin level requires excellent observational sensitivity. Optical elements in small-aperture telescopes such as BICEP Array are designed to optimize throughput and to minimize losses. My work in millimeter-wave instrumentation has concentrated on improving the performance of the optical chain: a series of vacuum windows, lenses, and infrared filters, each made from a different radio-transparent material with its own anti-reflection coating. Left: Schematic ray-trace and optical elements of a receiver in BICEP Array. The radio-transmissive multi-layer insulation (RT-MLI) is an infrared filter stack consisting of 12 layers of low-index polyethylene foam. The design includes an alumina infrared filter held at 50 K and a nylon filter at 4 K. For reference, the lens and window diameters are 638 mm and 876 mm, respectively.