The program For Inspiration and Recognition of Science and Technology (FIRST) for young students incorporates project-based learning (PBL) with designing and building wireless-controlled robots. The students are guided by experts, mostly engineers. The FIRST organization determines the theme of the robot annual competition. The goal of this research is to characterize and evaluate the effect of the FIRST program on graduates’ self-efficacy, interpersonal skills, and career choices in science, technology, engineering, and mathematics (STEM). The research participants included 297 FIRST graduates, mostly high schoolers, who responded to questionnaires, and five of them were interviewed. Analysis of the data showed that the FIRST program improved graduates’ interpersonal skills such as time management, teamwork skills, and self-efficacy, as well as had an impact on the graduates’ STEM career choices. The main factors impacting the graduates’ career choice was their exposure to robotics and to experts from the industry. The theoretical contribution is to the social cognitive theory (SCT) in the context of the FIRST program. Our study explains students’ career choice through correlations among students’ aspirations for choosing a career, their self-efficacy, their interpersonal skills, and their actual choice. The practical contribution lies in better understanding the robotic PBL program and expanding the STEM work force.

Embedded machine learning, TinyML, is a relatively new and fast-growing field of ML, enabling on-device sensor data analytics at low power requirements. This paper presents possible improvements to GMOS, a gas sensor, using TinyML technology. GMOS is a low-cost catalytic gas sensor, fabricated with the standard CMOS-SOI process, based on a suspended thermal transistor MOS (TMOS). Exothermic combustion reactions lead to temperature increases, which modify the suspended transistor’s (used as the sensing element) current-voltage characteristics. We were able to use GMOS measurements for gas classification (both for gas types, as well as concentration), resulting in high-proficiency gas detection at a low cost. Our preliminary results show great successes in the detection of ethanol and acetone gases. Moreover, we believe the method could be generalized to more gas types, concentrations, and gas mixes in future research.