Radio interferometers have the ability to precisely localize and better characterize the properties of sources. This ability is having a powerful impact on the study of fast radio transients, where a few milliseconds of data is enough to pinpoint a source at cosmological distances. However, recording interferometric data at millisecond cadence produces a terabyte-per-hour data stream that strains networks, computing systems, and archives. This challenge mirrors that of other domains of science, where the science scope is limited by the computational architecture as much as the physical processes at play. Here, we present a solution to this problem in the context of radio transients: realfast, a commensal, fast transient search system at the Jansky Very Large Array. Realfast uses a novel architecture to distribute fast-sampled interferometric data to a 32-node, 64-GPU cluster for real-time imaging and transient detection. By detecting transients in situ, we can trigger the recording of data for those rare, brief instants when the event occurs and reduce the recorded data volume by a factor of 1000. This makes it possible to commensally search a data stream that would otherwise be impossible to record. This system will search for millisecond transients in more than 1000 hours of data per year, potentially localizing several Fast Radio Bursts, pulsars, and other sources of impulsive radio emission. We describe the science scope for realfast, the system design, expected outcomes, and ways real-time analysis can help in other fields of astrophysics.

We present the first blind interferometric detection and imaging of a millisecond radio transient with an observation of transient pulsar J0628+0909. We developed a special observing mode of the Karl G. Jansky Very Large Array (VLA) to produce correlated data products (i.e., visibilities and images) on a time scale of 10 ms. Correlated data effectively produce thousands of beams on the sky that can localize sources anywhere over a wide field of view. We used this new observing mode to find and image pulses from the rotating radio transient (RRAT) J0628+0909, improving its localization by two orders of magnitude. Since the location of the RRAT was only approximately known when first observed, we searched for transients using a wide-field detection algorithm based on the bispectrum, an interferometric closure quantity. Over 16 minutes of observing, this algorithm detected one transient offset roughly 1' from its nominal location; this allowed us to image the RRAT to localize it with an accuracy of 1.6". With a priori knowledge of the RRAT location, a traditional beamforming search of the same data found two, lower significance pulses. The refined RRAT position excludes all potential multiwavelength counterparts, limiting its optical luminosity to L_i'<1.1x10^31 erg/s and excluding its association with a young, luminous neutron star.