September 2, 2026· ~6 min readWhat AI Reads Before It Recommends: Inside 7,534 Perplexity Citations Two Perplexity model variants were asked to recommend the five best tools in each of 380 B2B software categories. Their number-one picks matched in *
August 25, 2026· ~5 min readGraph Construction from Text with LLMs: The Bottleneck Hiding in Your GraphRAG Stack GraphRAG post-mortems almost always blame retrieval. They are looking in the wrong place. The accuracy ceiling in a graph-augmented system is set *up
July 30, 2026· ~6 min readKernel Forge: An Agent Harness for LLM-Based CUDA Kernel Generation and Optimisation > **Teaser:** A July 2026 paper introduces Kernel Forge — an agent harness that generates CUDA kernels from natural language descriptions, then itera
July 27, 2026· ~9 min readZot vs Pi vs OpenCode: A Hard Look at AI Coding Agent Harnesses<img src="/images/diagrams/zot-vs-pi-vs-opencode-benchmark.svg" alt="Benchmark comparison of zot, pi, and opencode showing startup time, memory usage,
July 18, 2026· ~4 min readNanbeige 4.2: Looped Transformer Architecture for Parameter-Efficient Model Capacity # Nanbeige 4.2: Looped Transformer Architecture On 22 July 2026, Nanbeige released Nanbeige 4.2, a 3-billion-parameter language model built on a Loo
July 16, 2026· ~4 min readMacaron-V1: Mixture-of-LoRA for Agentic AI at 748B Parameters # Macaron-V1: Mixture-of-LoRA for Agentic AI On 21 July 2026, Mind Lab released Macaron-V1, the first official release of their Mixture-of-LoRA (MoL