Memro is a persistent-memory system for AI coding agents, intended to let Claude Code, Cursor, and Codex retain useful information between sessions. It offers a hosted connection through a short device-login workflow or can be self-hosted with Docker Compose. Its Go API stores memories in PostgreSQL 16 with pgvector; retrieval combines vector similarity and full-text search using Reciprocal Rank Fusion. Agents can authenticate with Ed25519 identities whose private keys stay on the local machine, or with bearer tokens. The project also provides a CLI, an MCP server, and Python, JavaScript/TypeScript, and Go SDKs, alongside runnable integration examples.
Memories can be categorized as facts, events, procedures, or agent profiles, and the system supports time-aware search, tracking how information changes, identifying and resolving contradictions, and tracing relationships and provenance through reasoning graphs. The maintainers describe evaluations covering 500 LongMemEval questions, as well as tests for stale or conflicting facts, distractors, and identity boundaries. Self-hosting requires configuring the server and database; OpenAI credentials are needed for hybrid search, and a metrics token can protect monitoring endpoints. The code is available under the PolyForm Noncommercial License: personal, research, and testing use is allowed, while commercial or production use requires a Memro Cloud subscription or a separate commercial license.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
A Hacker News poster asks which practical AI skills individuals can learn online to benefit economically from LLMs and agents. They want specific, hands-on recommendations suited to someone with a moderately technical background, without requiring a degree. (news.ycombinator.com)
Former OpenAI researcher Will Depue claims hidden initialization patterns in six Chinese open-weight AI models match Nvidia Blackwell hardware, including systems restricted from export to China. He says the matches were reproduced on real chips and independently checked by researchers, though the claims are based on his forensic analysis. (twitter.com)
Antithesis built a system that beats NetHack on nearly 80% of game seeds within five minutes, across character builds. Successful runs take a median of under two minutes, advancing the company’s goal of creating smarter fuzzers and autonomous systems. (antithesis.com)
OpenCyclone is a free, open-source Android agent that uses your existing apps to carry out tasks through a hidden screen. It asks for approval before actions like sending messages and lets users connect their own AI models. (opencyclone.vercel.app)
Sam Bankman-Fried’s $500 million investment gave Alameda nearly 14% of Anthropic, a stake that would be worth about $37 billion at the AI company’s expected $1.5 trillion IPO valuation. The stake was sold in bankruptcy to repay FTX customers and creditors after Bankman-Fried’s businesses collapsed. (ft.com)
llama.cpp and ggml have reached v1.0, establishing stable public C APIs and clear versioning rules for breaking changes. The release strengthens the foundation for running local, multimodal AI across a wide range of hardware. (huggingface.co)
Today's best Hacker News stories, summarized and screenshotted, one email a day.