Show HN: Tried some experiments with architecture for Long term memory for LLM
π§ MindCache An open-source long-term memory engine for LLM agents. As conversations grow, important information gets buried. An AI needs to know what to remember, what has changed, which decisions still matter, and what is relevant now.
- βͺπ§ MindCache An open-source long-term memory engine for LLM agents.
- βͺAs conversations grow, important information gets buried.
- βͺAn AI needs to know what to remember, what has changed, which decisions still matter, and what is relevant now.
Hacker News (AI / LLM) files mainly under ai. We currently carry 4,672 of its stories.
Story provenance
Source Β· retrieval Β· rights Β· ranking β open for full record
inspect β
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | GitHub |
| Canonical URL | https://github.com/faisalhussain-devs/MindCache/tree/collapsed_tree |
| Publication time | Wed, 12 Aug 2026 20:03:11 +0000 |
| Retrieval time | 2026-08-12T20:06:33.943Z |
| Last seen | 2026-08-12T20:06:34.334Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch Β· cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | gQ7NoV1fJOs_ Β· 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes β open original |
| Substitutes article? | No β link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
π§ MindCache An open-source long-term memory engine for LLM agents. Agents forget. More history doesn't mean better memory. As conversations grow, important information gets buried. An AI needs to know what to remember, what has changed, which decisions still matter, and what is relevant now. Most systems treat this as a search problem. MindCache treats it as a memory problem. It turns conversations into organized, persistent memory and continuously updates that memory as the conversation evolves. π Read the 4-Minute Blog | π₯ Watch Demo Video | β‘ Quick Start | π Read the Engineering Journal π Want the technical deep dive? The short article explains the problem, architecture, and lessons behind MindCache in about four minutes.
β¦
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.