Children with Autism Spectrum Disorder (ASD) often experience challenges in logical reasoning and problem-solving, which can hinder their computational thinking (CT) development. This study aimed to examine the effectiveness of programming algorithm-based learning media in enhancing CT skills among children with ASD during Phase C informatics learning. A mixed-method approach combining Design-Based Research (DBR) and Single Subject Research (SSR) was employed. The study involved three children with ASD enrolled at SD BPI Bandung. The DBR process consisted of four stages: needs analysis, media design, development, and implementation. The intervention was conducted through an SSR design with three phases—Baseline 1 (A1), Intervention (B), and Baseline 2 (A2)—to assess changes in CT performance. All participants demonstrated notable improvements in CT skills following the intervention. For example, Participant KA’s score increased from 20% to 60%, KZ’s from 30% to 60%, and SB’s from 20% to 45%. The results indicated consistent upward trends and stable retention during the post-intervention baseline phase, suggesting that the programming algorithm learning media effectively enhanced computational thinking abilities. The findings support the use of structured, technology-assisted learning to promote CT in children with ASD. However, the small sample size limits generalizability, and future studies should include larger, more diverse participants. The study underscores the importance of developing adaptive, inclusive learning media that accommodate individual cognitive and sensory needs.