Xaira Therapeutics is introducing X-Automate, an internal laboratory system that puts vision-guided robots and agentic software on instruments built for people. Design work at the company can propose more protein and antibody candidates than a staffed lab can characterize, and many analytical machines still need someone to open a door, load a plate, or start a run. X-Automate pairs automation workcells with X-Bot, a six-degree-of-freedom arm built in-house on low-cost commercial hardware, and with X-Scientist, the agentic platform Xaira described two days earlier. A vision-language model reads the scene, chooses the next operation, and coordinates the handoff, including through the same on-screen controls a scientist would use. In a first deployment, the system connected a protein-production workcell to HPLC. Over seven days it retrieved, transported, and analyzed 18 96-well plates. Xaira reports that was 80 percent more than a modeled weekday 9-to-5 loading schedule, with the HPLC in use 97.2 percent of the time versus 53.6 percent in that model. A mobile robot moved plates between stations. X-Bot opened the drawer, swapped plates, and closed it, correcting its approach from camera feedback when the tray shifted. The week was not fully hands-off. Four interruptions in upstream plate storage needed manual recovery, though the company reported no errors in X-Bot or X-Scientist. It points to the same pattern for later bottlenecks, including flow cytometry, nanoDSF stability reads, and ELISA binding assays, and describes a later lab of about 100 instrument-level arms working beside scientists. Xaira came out of stealth in April 2024 with more than $1 billion in committed capital, incubated by ARCH Venture Partners and Foresite Labs, to apply AI across target selection, molecule design, and development. Scientific co-founder David Baker shared the 2024 Nobel Prize in Chemistry for computational protein design. Marc Tessier-Lavigne, former chief scientific officer of Genentech, is CEO. X-Automate is the physical layer next to X-Cell, X-Design, and X-Patient, meant to return experimental measurements to the models that propose the next candidates. Chief AI Scientist Bo Wang, a University of Toronto professor, announced the system.