Carbon dioxide (CO₂) detection is vital in various fields, such as environmental monitoring, healthcare, and industry. Metal oxides are the sensing material because of their high sensitivity and stability. However, they have limitations in detecting CO₂ as CO₂ is a chemically stable gas, which can be mitigated by post-treatment such as annealing or doping noble metals. Spark ablation, a versatile technique for synthesizing nanoparticles with controlled size and composition, was utilized in this study to produce SnO₂ nanoparticles, which were subsequently integrated into a specifically designed interdigitated electrode (IDE) structure for CO₂ sensing. The IDE configuration featured a fixed finger width of 5 μm, with the gap width ranging from 2 to 15 μm, and devices were fabricated with SnO₂ sensing areas of 1×1 cm and 1×4 cm. Noble metals such as Ag and Ru were decorated on the SnO₂ layer to enhance the catalytic activity and promote the CO₂ adsorption reaction. Surface densities of Ag and Ru under various printing conditions were analyzed using spark ablation. Based on these results, decorated SnO₂ layers were fabricated with surface coverages of 19% Ag and 12% Ru, respectively. Annealing was performed later at 450°C and 600°C, and XRD results revealed that the grain size of SnO₂ increased from 7 nm at 450°C to 12 nm at 600°C. Gas tests conducted in an N₂ atmosphere demonstrated CO₂ detection at 200°C, with a higher response observed for samples with Ru nanoparticle decoration, approximately twice that of the undecorated sample. However, no recovery was observed. Overall, samples annealed at 450°C exhibited higher responses compared to the ones annealed at 600°C. In an N₂/O₂ atmosphere, the sensor exhibited almost no response to CO₂. Further investigation revealed a response to O₂, suggesting that the low operating temperature may hinder the reaction between CO₂ and adsorbed oxygen species, failing CO₂ sensing. An anomalous p-conductive behaviour was also observed in response to O₂. Experimental results showed no correlation between device area, gap width, and sensing performance.