The welded joints of AISI 904 L super austenitic stainless steel are extensively utilized in automotive, ship, and aerospace industries, with joint quality assessed based on weld seam geometry, microstructures, and mechanical properties. However, selecting optimal weld input parameters to achieve desired specifications is traditionally time-consuming and relies heavily on the expertise of welding engineers or operators. To address this challenge and improve weld seam geometry and joint quality, this research proposes employing Long Short Term Memory (LSTM) modelling for laser welding of AISI 904 L stainless steel. The study systematically develops and evaluates LSTM models to predict weld bead geometry, including depth of penetration (DP), bead width (BW), and tensile strength (TS), with Argon shielding gases. By enhancing operator's expertise in parameter selection, this research significantly reduces time consumption, serving as a valuable tool for optimizing the welding process and enabling informed decision-making.