This paper presents a novel approach to fault-tolerant quantum computing called "algorithmic fault tolerance," which allows for constant-time logical operations with a broad class of quantum error-correcting (QEC) codes, including the surface code. The key idea is to use transversal operations and correlated decoding to achieve fault tolerance without requiring multiple syndrome extraction (SE) rounds, which is typically needed for fault-tolerant computation. The authors demonstrate that by leveraging the deterministic propagation of errors through transversal Clifford circuits and using partial syndrome information, the deviation from the ideal measurement result distribution can be made exponentially small in the code distance. This approach reduces the space-time cost of practical fault-tolerant quantum computation by orders of magnitude. The authors show that their approach can be implemented with a single SE round per logical operation, allowing for constant-time implementations of logical operations. They also demonstrate that this approach is competitive in performance, with numerical simulations showing clear threshold behavior and exponential suppression of logical error rates. The results are validated through a combination of proofs and simulations, including a simulation of state distillation factories, which show very little change to physical error thresholds. The paper also discusses the implications of their results for quantum computing, including the potential for significant practical savings over existing schemes. The authors highlight the importance of their work in the context of quantum computing, as it provides a theoretical foundation for algorithmic fault tolerance and demonstrates its applicability to universal quantum computing. The results are supported by numerical simulations and theoretical analysis, showing that the approach is both efficient and effective in reducing the space-time cost of fault-tolerant quantum computation.This paper presents a novel approach to fault-tolerant quantum computing called "algorithmic fault tolerance," which allows for constant-time logical operations with a broad class of quantum error-correcting (QEC) codes, including the surface code. The key idea is to use transversal operations and correlated decoding to achieve fault tolerance without requiring multiple syndrome extraction (SE) rounds, which is typically needed for fault-tolerant computation. The authors demonstrate that by leveraging the deterministic propagation of errors through transversal Clifford circuits and using partial syndrome information, the deviation from the ideal measurement result distribution can be made exponentially small in the code distance. This approach reduces the space-time cost of practical fault-tolerant quantum computation by orders of magnitude. The authors show that their approach can be implemented with a single SE round per logical operation, allowing for constant-time implementations of logical operations. They also demonstrate that this approach is competitive in performance, with numerical simulations showing clear threshold behavior and exponential suppression of logical error rates. The results are validated through a combination of proofs and simulations, including a simulation of state distillation factories, which show very little change to physical error thresholds. The paper also discusses the implications of their results for quantum computing, including the potential for significant practical savings over existing schemes. The authors highlight the importance of their work in the context of quantum computing, as it provides a theoretical foundation for algorithmic fault tolerance and demonstrates its applicability to universal quantum computing. The results are supported by numerical simulations and theoretical analysis, showing that the approach is both efficient and effective in reducing the space-time cost of fault-tolerant quantum computation.