Abstract Sophisticated infrared detection technology, operating through atmospheric transmission windows (usually between 3 and 5 μm and 8–13 μm), can detect an object by capturing its emitted thermal radiation, posing a threat to the survival of targeted objects. As per Wien’s displacement law, the shift of peak wavelength towards shorter wavelengths as blackbody temperature rises, underscores the significance of the 3–5 μm range for ultra-high temperature objects (e.g., at 400 °C), emphasizing the crucial need to control this radiation for the objects’ viability. Additionally, effective heat management is essential for ensuring the consistent operation of these ultrahot entities. In this study, based on a database with high-temperature resist materials, we introduced a material-informatics-based framework aimed at achieving the inverse design of simultaneous thermal camouflage (low emittance in the 3–5 μm range) and radiative cooling (high emittance in the non-atmospheric window 5–8 μm range) tailored for ultrahigh-temperature objects. Utilizing the transfer matrix method to calculate spectral properties and employing the particle swarm optimization algorithm, two optimized multilayer structures with desired spectral characteristics are obtained. The resulted structures demonstrate effective infrared camouflage at temperatures up to 250 °C and 500 °C, achieving reductions of 86.7 % and 63.7 % in the infrared signal, respectively. At equivalent heating power densities applied to the structure and aluminum, structure 1 demonstrates a temperature reduction of 29.4 °C at 0.75 W/cm2, while structure 2 attains a temperature reduction of 57.5 °C at 1.50 W/cm2 compared to aluminum, showcasing enhanced radiative cooling effects. This approach paves the way for attenuating infrared signals from ultrahigh-temperature objects and effectively managing their thermal conditions.