**Abstract:** A significant advancement in the field of spectroscopy has been reported by Professor Yongfeng Mei's research group at Fudan University's Department of Materials Science and the International Institute of Smart Nano Robots and Nano Systems. Their work, published on August 13, 2024, in the *Proceedings of the National Academy of Sciences of the United States of America* (PNAS), introduces a novel design for a miniaturized reconstructive spectrometer that combines the advantages of traditional and computational spectrometers. Dr. Chunyu You is the first author, and Professor Yongfeng Mei is the corresponding author. The research was highlighted as the cover story of the PNAS issue. The new design integrates self-referencing Fabry-Perot resonators, enabling artificial intelligence (AI) algorithms to search for both spectral and algorithmic parameters in a higher-dimensional parameter space. This integration allows for more accurate and stable spectral reconstructions without the need for manual calibration, which is a common issue in previously reported reconstructive miniature spectrometers. The spectrometer can be manufactured at the wafer level using mature integrated circuit (IC) processes, ensuring scalability and cost-effectiveness. The device measures millimeters in size, making it suitable for a wide range of applications that require miniaturized spectroscopic testing. The team's spectrometer demonstrates excellent performance across the visible light spectrum (400-800 nm), achieving a resolution of approximately 2.5 nm, an average wavelength deviation of about 0.27 nm, a resolution power of up to 5806, and a resolution-to-bandwidth ratio of about 0.46%. These metrics are comparable to those of commercial fiber optic spectrometers but with significantly reduced costs and size, marking a significant step forward in the miniaturization of spectroscopic systems. To showcase the practical utility of their design, the researchers integrated the adaptive miniature spectrometer with microfluidic and mechanical scanning systems, demonstrating its effectiveness in common laboratory applications such as transmission, absorption, and photoluminescence spectroscopy. The results obtained from these tests were consistent with those from commercial fiber optic spectrometers, validating the device's accuracy and reliability. Additionally, the team demonstrated the spectrometer's potential in hyperspectral imaging, laying the groundwork for its use as individual pixels in high-resolution cameras. The integration of self-referencing narrowband filtering channels is a key innovation. These channels provide a set of low-resolution but highly accurate reference signals at fixed wavelengths, which guide and correct the AI-reconstructed spectral signals. This feature ensures that the spectrometer can deliver high-resolution and stable results without the need for manual intervention, addressing a critical limitation in previous reconstructive miniature spectrometers. The research is supported by the National Key Research and Development Program, the National Natural Science Foundation of China, and the Shanghai Science and Technology Commission. Experiments were conducted in Fudan University's Micro-Nano Fabrication and Device Public Laboratory. In conclusion, this study offers a new approach to developing miniature reconstructive spectrometers with universal applicability and high robustness. By leveraging established CMOS IC manufacturing processes, the spectrometer can be integrated into complementary metal-oxide-semiconductor (CIS) image sensors, opening up opportunities for its use in mobile and portable measurements, automotive machine vision, and distributed monitoring systems. This development has the potential to revolutionize the way spectroscopic data is collected and analyzed, making advanced spectroscopy more accessible and cost-effective. **Key Elements:** - **Key Event:** Publication of a novel design for a miniaturized reconstructive spectrometer. - **People Involved:** Professor Yongfeng Mei, Dr. Chunyu You, and the research team at Fudan University. - **Location:** Fudan University, Shanghai, China. - **Time:** August 13, 2024. - **Publication:** *Proceedings of the National Academy of Sciences of the United States of America* (PNAS). - **Technology:** Integration of self-referencing Fabry-Perot resonators with AI algorithms. - **Applications:** Machine vision, environmental monitoring, biomedical applications, and hyperspectral imaging. - **Performance Metrics:** Resolution of 2.5 nm, average wavelength deviation of 0.27 nm, resolution power of 5806, and resolution-to-bandwidth ratio of 0.46%. - **Significance:** Potential to reduce costs and size while maintaining performance comparable to commercial spectrometers, enabling broader and more cost-effective use in various fields.