Feed aggregator

Article URL: https://github.com/ggml-org/llama.cpp/discussions/22891

Comments URL: https://news.ycombinator.com/item?id=48080800

Points: 2

# Comments: 0

Article URL: https://github.com/nooga/xsofy

Comments URL: https://news.ycombinator.com/item?id=48080755

Points: 3

# Comments: 0

Article URL: https://blog.google/innovation-and-ai/technology/developers-tools/expanded-gemini-api-file-search-multimodal-rag/

Comments URL: https://news.ycombinator.com/item?id=48080702

Points: 7

# Comments: 0

Article URL: https://codebrainery.com/articles/snyk-claude-code-real-time-security-scanning-for-ai-code

Comments URL: https://news.ycombinator.com/item?id=48080697

Points: 2

# Comments: 0

Article URL: https://www.cureus.com/articles/488870-closed-loop-extracorporeal-vascular-cleaning-by-staged-chemical-dissolution

Comments URL: https://news.ycombinator.com/item?id=48080684

Points: 2

# Comments: 0

Article URL: https://github.com/narcotic-sh/modafinil

Comments URL: https://news.ycombinator.com/item?id=48080675

Points: 3

# Comments: 0

Article URL: https://www.youtube.com/watch?v=p6Ejmhwb8Sc

Comments URL: https://news.ycombinator.com/item?id=48080668

Points: 2

# Comments: 0

Article URL: https://github.com/aniongithub/devcontainer-mcp

Comments URL: https://news.ycombinator.com/item?id=48080662

Points: 3

# Comments: 0

Article URL: https://surfingcomplexity.blog/2026/05/09/flipping-the-bozo-bit-on-flips-the-learning-off/

Comments URL: https://news.ycombinator.com/item?id=48080658

Points: 3

# Comments: 0

Article URL: https://www.digitalocean.com/blog/powering-the-inference-era

Comments URL: https://news.ycombinator.com/item?id=48080656

Points: 2

# Comments: 0

Article URL: https://clereviewofbooks.com/the-laboratory-and-the-artist/

Comments URL: https://news.ycombinator.com/item?id=48080647

Points: 3

# Comments: 0

Article URL: https://physicsworld.com/a/memory-device-breaks-high-temperature-performance-record/

Comments URL: https://news.ycombinator.com/item?id=48080645

Points: 4

# Comments: 0

Article URL: https://www.jpost.com/opinion/article-895285

Comments URL: https://news.ycombinator.com/item?id=48080627

Points: 5

# Comments: 2

Article URL: https://exquisite.tube/w/dkV6kWiT9sp2y6xVwkH1iF

Comments URL: https://news.ycombinator.com/item?id=48080625

Points: 3

# Comments: 0

Article URL: https://github.com/aabiji/kepler

Comments URL: https://news.ycombinator.com/item?id=48080609

Points: 2

# Comments: 0

Semantic search alone is not enough to capture all connected facts, they will capture the semantically most identical memory only.

Tested on HotpotQA public dataset:

Vector + BM25 + entity graph: BothFound@5 71.5% Vector + BM25 only: BothFound@5 59.5%

Entity graph is the game changer to extract connected facts.

More Benchmark result:

LongMemEval-S: 84.8% recallAll@5 LoCoMo-10: 59% vs zep cloud 28%

What is your approach for connected facts retrieval ?

Comments URL: https://news.ycombinator.com/item?id=48080597

Points: 1

# Comments: 0

Article URL: https://www.sciencealert.com/plants-can-sense-the-sound-of-rain

Comments URL: https://news.ycombinator.com/item?id=48080593

Points: 1

# Comments: 0

This is ymawky, a static file web server for MacOS written entirely in ARM64 assembly. It supports GET, PUT, DELETE, HEAD, and OPTIONS requests, and supports Range: bytes=X-Y headers (which allows scrubbing for video streaming). It decodes percent-encoded URLs, strictly enforces docroot, serves custom error pages for any HTTP error response, supports directory listing, and has (some) mitigations against slowloris-like attacks.

I’ve also written a more detailed writeup here: https://imtomt.github.io/ymawky/

Comments URL: https://news.ycombinator.com/item?id=48080587

Points: 3

# Comments: 0

Article URL: https://www.ft.com/content/a40cafcc-0fa4-4e70-9e24-90d826aea56d

Comments URL: https://news.ycombinator.com/item?id=48080540

Points: 1

# Comments: 0

some of the friction of using coding agents for product building (not just writing code) is every new session starts from scratch.

draft is my attempt at a fix. before getting into how it works — a lot of memory/context plugins solve this by calling an external api or running their own model, which either charges you separately or quietly inflates your usage. draft runs entirely within your existing claude subscription.

it's a codex/claude code plugin a few core pieces:

1. session-init hook

on every session start, a hook fires and injects a structured context summary into claude's working memory. specifically, it reads the `name` and `description` frontmatter from each context index file — company, product, priorities, team, decisions — and assembles them into a single compressed summary under ~5k tokens.

2. learning — inference-driven and manual

learning is inference-driven rather than hook-based. the main claude agent reads its instructions and decides when something is worth persisting — a decision made, a direction abandoned — and calls `/draft:learn` as a subagent to write it back to the appropriate context file. there's no session-end hook or background process; the agent is exercising judgment about what's durable. you can also invoke `/draft:learn` manually at any time.

over time, claude accumulates enough context to flag contradictions — if you go in a direction that conflicts with a decision you made two weeks ago, it'll surface it. that's been the most unexpectedly useful part for me as a solo founder.

3. cross-agent architecture

the workspace lives at `~/.draft/workspace/`, claude code and cursor share the workspace natively. codex requires a separate install (curl setup script, since there's no plugin marketplace yet) but reads from the same files.

---

would love if folks tried it out, cloned it and ran locally or gave me some examples of their coding agent setups!

its 100% free and open source. mit licensed. no account, no api key.

Comments URL: https://news.ycombinator.com/item?id=48080538

Points: 1

# Comments: 0