Topic
Retrieval, context and what earns a place in the window
Retrieval is usually described as a search problem and behaves like an editing problem. The hard part is not finding relevant text, it is deciding what deserves the context window and what is actively making the answer worse. Work on retrieval systems, context engineering and the tooling underneath them.
26 pieces
Teardowns
8- When a Knowledge Graph Beats a Vector Database Vector search retrieves what looks similar. It cannot tell you which shipped assemblies a bad plating bath touches two hops out. That is traversal, and traversal is a different operation.
- Genome Language Models: What an Engineer Should Actually Take From Evo 2 DNA is a sequence and transformers are good at sequences. That analogy carries genome language models further than it should. What Evo 2 actually does, what it is used for and where the pitch breaks.
- An AI research assistant for literature review does not read for you Gemini Notebook answers only from sources you supply. That fence moves the error rather than deleting it, and the part it cannot do, judging whether your corpus is any good, stays with you.
- The AI Data Privacy Checklist I Run Before Shipping a Feature Retention windows, zero-retention gaps, personal data in vector stores and the erasure problem, written for the engineer who has to configure it before Friday.
- Context Engineering Is Deciding What Earns a Place in the Window Windows grew to a million tokens and the industry assumed retrieval was solved. Attention still degrades, cost still scales, and everything you paste competes for the model's attention.
- How DeepSeek Builds World-Class AI on a Shoestring Budget The real story behind the $5.6 million model that panicked Silicon Valley, explained so you actually understand it.
- Every insert writes a part Manufacturing telemetry is append-heavy, high-cardinality and almost always queried as a time-bounded aggregate. A columnar store fits that shape exactly, and breaks on updates, joins and small insert
- Context Engineering vs Prompt Engineering: One Task, Two Fixes, Two Bills Prompt engineering optimises the wording. Context engineering decides what is in the window at all. As windows grew the bottleneck moved from phrasing to selection, and the bill moved with it.
Tools
12 of 18- ChatGPT review OpenAI's assistant and API, now on the GPT-5.6 Sol, Terra and Luna tiers, with a 1M token context window on all three and reasoning effort you set per request.
- Databricks Mosaic AI review The AI layer inside Databricks: Agent Bricks, model serving, vector search and MLflow, all resolving back to Unity Catalog governed data.
- Dify review Open-source LLM app platform from LangGenius for agentic workflows and RAG pipelines. Modified Apache 2.0: self-host freely, multi-tenant SaaS needs a commercial licence.
- FiftyOne review Apache 2.0 Python library and app for curating vision datasets. Query and slice samples, explore embeddings, find label mistakes and push fixes back to CVAT or Label Studio.
- Milvus review Apache 2.0 vector database from Zilliz, self-hostable from a pip install up to a billion-vector cluster. The open engine under Zilliz Cloud, minus the proprietary Cardinal search core.
- RAGFlow review InfiniFlow's Apache 2.0 RAG engine built on deep document parsing. DeepDoc handles layout, tables and OCR before chunking, and every citation points back to a highlighted region in the source PDF.
- Zilliz Cloud review Managed vector database built on Milvus, the open-source engine behind most billion-scale retrieval systems, for teams who need hybrid search without running the cluster themselves.
- pgvector review Postgres extension that adds vector types and approximate nearest neighbour indexes, so embeddings live in the same database as the rest of your data.
- Kimi / Moonshot AI (月之暗面) review Moonshot AI’s long-context AI assistant, pioneering ultra-long document processing with a 2 million token context window.
- Quark / 夸克 review Alibaba’s AI-powered search and productivity app combining intelligent search, document tools, cloud storage, and AI assistant capabilities.
- Apache Doris review MPP analytics database that began as Baidu's Palo. Version 4.0 put HNSW vector indexes, BM25 full-text search and SQL-callable LLM functions into the same engine that runs the OLAP queries.
- Qdrant review Rust vector search engine with native hybrid dense and sparse retrieval, self-hostable under Apache 2.0 or run as a managed cloud cluster.