High-resolution spectroscopy is a powerful tool to characterize the exoplanet atmosphere as it is able to resolve molecular bands into individual absorption lines allowing a robust molecular species detection. This technique is able to distinguish the exoplanet lines from telluric and stellar lines due to the variation of Doppler shift of the planet during the observation. It has been predicted that titanium oxide (TiO) and/or vanadium oxide (VO) causing thermal inversions in the atmosphere of the very hot Jupiter but there was no strong observational evidence to support with. Thermal inversion is a common phenomenon that can be found in the solar system planets including our Earth. The existence of this layer affects the atmospheric dynamics and the thermal evolution of the planet. The hot Jupiter is a very unique planet that we do not have in our solar system. Therefore, it is very important to detect the responsible compound of thermal inversion in the hot Jupiter to understand the exoclimate of this planetary group. The aims of this thesis are to find TiO in the atmosphere of hot Jupiter and solve whether it is the responsible molecule for the thermal inversion layer in the hot Jupiter or not using highresolution spectroscopy. I observed the emission spectrum ofWASP-33b (Tday= 3400 K) and the transmission spectrum of HD 209458b (Teq= 1800 K) using High Dispersion Spectrograph (HDS; R 165,000) on Subaru telescope in the wavelength range of 0.62-0.88 mm. We remove the systematics from the instrument, the telluric and stellar lines using SYSREM and cross-correlated with planet model spectrum. I succeeded to simultaneously detect the first high-resolution TiO emission signature and confirm the existence of the stratosphere on the day-side atmosphere of WASP-33b by 4.8-s. This detection provides the first direct evidence of TiO being the responsible molecule of thermal inversion in the hot Jupiter and estimates the maximum TiO condensate size of 3-5 mm in the atmosphere ofWASP-33b. Meanwhile, there is no TiO signature in the transmission spectrum of HD209458b down to the volume mixing ratio of 10?8 which is set by the enhanced Rayleigh scattering slope. The nondetection is best explained by the cold trap in its atmosphere and is consistent with the previous studies constraining the non-inverted atmosphere of HD 209458b. These results highlight the need to continue the research of finding TiO in the atmosphere of moderate hot Jupiter to extend our understanding of the thermal inversion layer in the hot Jupiter.