AI Infrastructure

AI Infrastructure

The Inference-Time Compute Infrastructure Frontier: Scaling the Serving Stack for DeepSeek-R1 and Reasoning Models

Deconstruct the physical infrastructure shift of inference-time compute. Learn how to scale the serving stack for reasoning models like DeepSeek-R1 and OpenAI o1, implement prefill-decode disaggregation, configure SGLang and vLLM parsers, and overcome critical production roadblocks such as JIT compilation deadlocks and MoE network saturation.

Lead Infrastructure ArchitectAugust 21, 2026

Read

Scaling the Inference Wall: Disaggregated Prefill-Decode, Multi-Head Latent Attention (MLA), and the vLLM V1 vs. SGLang Duel

The 2026 LLM Infrastructure Frontier: Native FP4, Weight-Absorbed MLA, and Disaggregated Prefill-Decode Serving

The Inference-Time Compute Infrastructure Frontier: Scaling the Serving Stack for DeepSeek-R1 and Reasoning Models

The 2026 LLM Infrastructure Frontier: SGLang vs. vLLM V1, MLA Compression, and EAGLE-3 Speculative Decoding