latent.fs / posts — 46 items

/feed.xml

Blog posts

2026-11 2 posts

  1. xsm: A Message Channel for Agent Collaboration

    Preserving existing work environments and agent autonomy while organizing collaboration that people can understand and intervene in.

    AI·Agent #45

  2. Weight-Space Programming

    How an LLM edited weights directly, and where it leads

    AI·Model #44

2026-10 10 posts

  1. What Happens to HNSW Recall After Adding Vectors? (1M+1M)

    HNSW recall in pgvector and Qdrant after adding one million vectors

    DB

  2. What Happens to HNSW Recall After Adding Vectors? (10K+10K)

    HNSW recall in pgvector and Qdrant after adding ten thousand vectors

    DB

  3. Minority Responses in Persona-Prompted LLMs: Following Architect of Silence

    Minority responses lost in persona-prompted LLM surveys

    Research #43

  4. How Managed Agents Takes an Agent Apart: Brain, Hands, and Session

    The session, harness, and sandbox split in Managed Agents, with a workshop run

    AI·Agent #42

  5. Model Providers Are Taking Over the Agent Runtime: Managed Agents and the OpenAI Agents API

    Runtime abstractions in Managed Agents and the OpenAI Agents API

    AI·Agent #41

  6. Using Claude Subscription Credits for the API

    Scope and use of the API credits in Max and Team plans

    AI·Infra #40

  7. In the Agent Era, the Scope a Specialist Software Supplier Can Choose

    What a specialist software supplier can sell in the agent era

    AI·Agent #39

  8. Seeds and Germination: Software That Evolves in the Field

    The structure of software that adapts and evolves at customer sites

    AI·Agent #38

  9. Every Organization Needs an Agent Runtime

    General-purpose agent runtimes in air-gapped organizations

    AI·Agent #37

  10. Separating the Worker from the Tool

    The separation of worker and tool in the age of general-purpose agents

    AI·Agent #36

2026-09 2 posts

  1. Can tgrep, CodeGraph, Mantic, or ripwire Replace ripgrep?

    Four code search tools against ripgrep on eight repositories

    AI·Agent #35

  2. ego-browser and Aside Fail Differently

    How ego-browser and Aside fail on the same browser tasks

    AI·Agent #34

2026-07 2 posts

  1. Driving a Free Colab GPU from the CLI

    A free Colab GPU driven from colab CLI, with a PaddleOCR-VL example

    AI·Infra #33

  2. Wiring codex to a Local Qwen — ollama, vLLM, and the developer role

    Connecting codex CLI to a local Qwen on ollama and vLLM

    AI·Agent #32

2026-06 1 post

  1. Speeding up LLM inference with MTP and diffusion

    Speed and accuracy of MTP and diffusion decoding on Gemma 4 and Qwen 3.6

    AI·Inference #31

2026-05 29 posts

  1. Requirements as Latent State

    Requirements as latent states and natural language as their observation

    AI·Agent #30

  2. Skill and Harness

    The relation between skills that compress behavior and harnesses that judge it

    AI·Agent #29

  3. Harness as Environment

    The harness as the condition for agent adaptation

    AI·Agent #28

  4. Memory as Adaptation

    Memory layers and policy adaptation in long-running agents

    AI·Agent #27

  5. Weights, Prompts, Codes as Parameters

    Weights, prompts, and code as three layers of parameters in one policy space

    AI·Agent #26

  6. Comparing Four LightRAG Variants — Same Root, Different Production Strategies

    A source-level comparison of three LightRAG derivatives

    AI·RAG #25

  7. PostgreSQL Lakehouse (2/2) — Distributed Processing and Citus Integration

    Distributed processing engines and Citus integration for a PostgreSQL lakehouse

    DB #24

  8. PostgreSQL Lakehouse (1/2) — When DuckLake Hit a Wall, pg_lake Was There

    DuckLake's libpq collision and a pg_lake-based lakehouse

    DB #23

  9. PG-Strom SSB Benchmark — Arrow FDW Comes Before GPU

    PG-Strom GPU acceleration and Arrow FDW on the SSB benchmark

    DB #22

  10. GraphDB Benchmark (2/2) — Workload Matrix and Final Recommendations

    Workload performance and selection criteria for eight graph engines

    AI·RAG #21

  11. PDF to Markdown — Five Tools Compared

    Five PDF-to-Markdown converters evaluated on an academic paper

    AI·RAG #17

  12. Markdown to Slides and PDF — A zsh Pipeline

    zsh functions for markdown-to-slides and PDF with pandoc and XeLaTeX

    Development #16

  13. Sharing Claude Code Sessions via Symlinked .jsonl

    Session sharing across Claude Code accounts through symlinks

    AI·Agent #15

  14. Apple Silicon LLM Inference — Five Backends Compared

    Single and concurrent request performance of five LLM backends on Apple Silicon

    AI·Infra #13

  15. LightRAG Without Apache AGE — Graph Storage in Recursive CTE

    LightRAG graph storage on plain PostgreSQL recursive CTEs

    AI·RAG #14

  16. Codex App Server Python SDK — JSON-RPC v2 over stdio

    Installation and call structure of a Python SDK for codex app-server

    AI·Agent #12

  17. Running Multiple Claude Code Accounts on One Machine

    Multiple Claude Code accounts with CLAUDE_CONFIG_DIR and a shell function

    AI·Agent #11

  18. Long-Context Evaluation — NIAH and Lost in the Middle

    The limits of NIAH and measurements of Lost in the Middle

    AI·Model #10

  19. Should Databases Handle AI Too?

    Five in-database AI systems and the abstraction boundary

    DB #9

  20. autocorrect.zsh — 167 lines of zsh that fix failed commands

    A zsh script that fixes failed commands with Gemini Flash and shell hooks

    Development #8

  21. Replacing Elasticsearch with PostgreSQL

    Korean hybrid search on PostgreSQL alone, compared with Elasticsearch

    DB #7

  22. NVIDIA NIM API — Free Inference for GLM, Kimi, Nemotron, and Gemma 4

    Usage and limits of the free NVIDIA NIM inference API

    AI·Infra #3

  23. GraphDB Benchmark, Eight Engines (Part 1) — Decomposing the RCTE 290x Gap

    The causes of the throughput gap between PostgreSQL RCTE and Apache AGE

    AI·RAG #20

  24. Disk Utilities Through Two Surfaces — zsh Aliases and a Claude Skill

    Disk utilities shared by zsh aliases and a Claude Skill

    Development #19

  25. Claude Code Settings Sync and Troubleshooting

    Settings sync and troubleshooting for multiple Claude Code accounts

    AI·Agent #18

  26. 10 OpenAI Models Through Quick Benchmarks — The Model Isn't as Smart as You Pay

    Accuracy of ten OpenAI models on three reasoning problems

    AI·Model #6

  27. GitHub Models Inference API — Free Model Access Tested

    Usage and limits of the free GitHub Models inference API

    AI·Infra #5

  28. Experimenting with the GPT-5 Responses API Web Search Tool

    Model-by-model behavior of the GPT-5 Responses API web search tool

    AI·Model #4

  29. Life As It Could Be

    A guide to Artificial Life through the ALife 2025 exhibition

    Research #1

TAGS 196

All 196 tags →