# TiDB

TiDB | SQL at Scale

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## How Vector Search Works Inside TiDB for AI Applications

DevFeed: [How Vector Search Works Inside TiDB for AI Applications](<https://devfeed.tech/articles/how-vector-search-works-inside-tidb-for-ai-applications-128968.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/integrating-vector-search-into-tidb-for-ai-applications/>)

Author: Xin Shi

Published: 2026-11-10T18:43:16Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [ann](<https://devfeed.tech/topics/ann.md>), [cosine similarity search](<https://devfeed.tech/topics/cosine-similarity-search.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [application](<https://devfeed.tech/tags/application.md>), [applications](<https://devfeed.tech/tags/applications.md>), [article](<https://devfeed.tech/tags/article.md>), [can](<https://devfeed.tech/tags/can.md>), [context](<https://devfeed.tech/tags/context.md>), [customer](<https://devfeed.tech/tags/customer.md>), [database](<https://devfeed.tech/tags/database.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [first](<https://devfeed.tech/tags/first.md>), [form](<https://devfeed.tech/tags/form.md>), [inside](<https://devfeed.tech/tags/inside.md>), [job](<https://devfeed.tech/tags/job.md>), [keep](<https://devfeed.tech/tags/keep.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [product](<https://devfeed.tech/tags/product.md>), [search](<https://devfeed.tech/tags/search.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tidb-serverless](<https://devfeed.tech/tags/tidb-serverless.md>), [vector](<https://devfeed.tech/tags/vector.md>), [vector-search](<https://devfeed.tech/tags/vector-search.md>), [works](<https://devfeed.tech/tags/works.md>)

### AI overview

TiDB stores embeddings with transactional data: TiKV commits writes, then TiFlash replicates them and serves HNSW vector search. This lets RAG retrieval join vectors with permissions, tenants, and document versions in one SQL query, and makes committed rows searchable without a separate sync pipeline. New or updated rows in TiFlash's Delta layer remain searchable but run more slowly until compaction moves them into the indexed Stable layer. The article reports HNSW recall generally above 90% and notes that a 500 MiB dataset of 768-dimensional vectors may take up to 20 minutes to index. Vector search requires TiFlash; pre-KNN WHERE filters bypass the index, so the suggested subquery pattern can return fewer than K rows. Availability varies by deployment: TiDB Cloud documentation labels the feature beta, while Self-Managed documentation calls it experimental and not recommended for production.

### Source excerpt

Key Takeaways When a RAG application keeps embeddings in a separate vector store, an edited support article can keep surfacing in its old form until the next sync job runs. The model then answers from stale context, and nobody notices until a customer does. Knowing how vector search works inside your database tells you how [...] The post How Vector Search Works Inside TiDB for AI Applications appeared first on TiDB.

## How to Build a LlamaIndex Vector Store With TiDB

DevFeed: [How to Build a LlamaIndex Vector Store With TiDB](<https://devfeed.tech/articles/how-to-build-a-llamaindex-vector-store-with-tidb-109548.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/llamaindex-vector-store/>)

Author: Akshata Hire

Published: 2026-10-30T19:16:02Z

Content type: tutorial

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [llamaindex](<https://devfeed.tech/topics/llamaindex.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [vector embeddings in AI](<https://devfeed.tech/topics/vector-embeddings-in-ai.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [hybrid-search](<https://devfeed.tech/topics/hybrid-search.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>)

Tags: [build](<https://devfeed.tech/tags/build.md>), [can](<https://devfeed.tech/tags/can.md>), [content](<https://devfeed.tech/tags/content.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [first](<https://devfeed.tech/tags/first.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [index](<https://devfeed.tech/tags/index.md>), [llamaindex](<https://devfeed.tech/tags/llamaindex.md>), [memory](<https://devfeed.tech/tags/memory.md>), [metadata](<https://devfeed.tech/tags/metadata.md>), [node](<https://devfeed.tech/tags/node.md>), [node-embeddings](<https://devfeed.tech/tags/node-embeddings.md>), [production](<https://devfeed.tech/tags/production.md>), [queries](<https://devfeed.tech/tags/queries.md>), [store](<https://devfeed.tech/tags/store.md>), [systems](<https://devfeed.tech/tags/systems.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tutorial](<https://devfeed.tech/tags/tutorial.md>), [vector](<https://devfeed.tech/tags/vector.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

TiDBVectorStore gives LlamaIndex a persistent, shared SQL-backed store for node embeddings, text, and JSON metadata; the existing VectorStoreIndex flow stays the same by passing it through StorageContext. The tutorial covers loading and chunking documents, matching vector_dimension to the embedding model, indexing and reconnecting, and filtering queries by fields such as tenant or document type. TiDB metadata filters run before distance ordering, returning exact filtered results but scanning matching rows, so cost grows with their number. TiDBVectorStore supports LlamaIndex's default query mode only; keyword or hybrid retrieval needs a separate path. TiDB full-text search uses BM25 but is limited to TiDB Cloud Starter in five AWS regions and does not guarantee exact phrase matching. The text also notes vector-search version requirements and public-preview status, and recommends evaluating top-k retrieval, planning tenant isolation, and tuning inserts for packet limits.

### Source excerpt

A LlamaIndex vector store is where your index keeps node embeddings, usually alongside the text chunks and metadata, so queries can find similar content. The default store lives in memory, so a restart or redeploy wipes the index and sends you back to your embedding provider to pay for the same corpus again. Production systems [...] The post How to Build a LlamaIndex Vector Store With TiDB appeared first on TiDB.

## TiDB Cloud BYOC Public Preview: Fully Managed Distributed SQL in Your AWS Account

DevFeed: [TiDB Cloud BYOC Public Preview: Fully Managed Distributed SQL in Your AWS Account](<https://devfeed.tech/articles/tidb-cloud-byoc-public-preview-fully-managed-distributed-sql-in-your-aws-account-88903.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/tidb-cloud-byoc-public-preview/>)

Author: Wendy Cheng

Published: 2026-10-16T20:02:03Z

Content type: release

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Redpanda BYOC](<https://devfeed.tech/topics/redpanda-byoc.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [Service fabric](<https://devfeed.tech/topics/service-fabric.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [aws](<https://devfeed.tech/tags/aws.md>), [backup](<https://devfeed.tech/tags/backup.md>), [byoc](<https://devfeed.tech/tags/byoc.md>), [cost-management](<https://devfeed.tech/tags/cost-management.md>), [data-residency](<https://devfeed.tech/tags/data-residency.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [product](<https://devfeed.tech/tags/product.md>), [production](<https://devfeed.tech/tags/production.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tidb-cloud](<https://devfeed.tech/tags/tidb-cloud.md>), [tidb-x](<https://devfeed.tech/tags/tidb-x.md>)

### AI overview

PingCAP announces the public preview of TiDB Cloud BYOC, which runs a managed distributed SQL database in a customer's AWS account. The service keeps database compute, storage, and networking in that account while TiDB Cloud manages deployment and operations. It includes scaling, workload isolation, backups, security controls, and AWS billing integration; pricing combines AWS infrastructure costs with PingCAP service and support fees.

### Source excerpt

Teams running distributed SQL in production face a trade-off. A managed service removes the work of upgrades, scaling, and backups, and it moves the database into the vendor's cloud account. Self-managing TiDB keeps the database in your account and leaves every upgrade, backup, and incident to your team. Today, we are announcing the public preview [...] The post TiDB Cloud BYOC Public Preview: Fully Managed Distributed SQL in Your AWS Account appeared first on TiDB.

## TiDB Cloud Announces Premium Availability Plans and Previews for BYOC, PostgreSQL and Agent Workspaces

DevFeed: [TiDB Cloud Announces Premium Availability Plans and Previews for BYOC, PostgreSQL and Agent Workspaces](<https://devfeed.tech/articles/different-starting-points-one-data-foundation-what-s-new-in-tidb-cloud-87683.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/tidb-cloud-scaile-2026-product-announcements/>)

Author: Brian Foster

Published: 2026-10-15T15:00:00Z

Content type: release

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Cloud](<https://devfeed.tech/topics/cloud.md>), [sql-database](<https://devfeed.tech/topics/sql-database.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [announcement](<https://devfeed.tech/tags/announcement.md>), [bring-your-own-cloud](<https://devfeed.tech/tags/bring-your-own-cloud.md>), [byoc](<https://devfeed.tech/tags/byoc.md>), [database](<https://devfeed.tech/tags/database.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [product](<https://devfeed.tech/tags/product.md>), [product-announcements](<https://devfeed.tech/tags/product-announcements.md>), [tidb-cloud](<https://devfeed.tech/tags/tidb-cloud.md>), [tidb-x](<https://devfeed.tech/tags/tidb-x.md>)

### AI overview

PingCAP announced TiDB Cloud Premium customer availability for November 2026 and public previews of BYOC, PostgreSQL support on Starter, and TiDB Cloud Filesystem for persistent AI agent workspaces. The PostgreSQL preview is limited to development and evaluation and has no production SLA. PingCAP also plans to open source TiDB X core code by the end of 2026.

### Source excerpt

Nobody approves a project called "add complexity." It builds up one shard, one region, and one compliance requirement at a time, until a single database has turned into a fleet and the team running it has grown to match. Atlassian reached that point with Forge SQL, its hosted relational database service for developers, which ran [...] The post Different Starting Points, One Data Foundation: What's New in TiDB Cloud appeared first on TiDB.

## Persistent AI Agent Memory Explained: How Agents Remember Across Sessions

DevFeed: [Persistent AI Agent Memory Explained: How Agents Remember Across Sessions](<https://devfeed.tech/articles/persistent-ai-agent-memory-explained-how-agents-remember-across-sessions-85897.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/persistent-ai-memory/>)

Author: M. Nivanya

Published: 2026-10-13T16:06:08Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [sessions](<https://devfeed.tech/topics/sessions.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [rocksdb](<https://devfeed.tech/topics/rocksdb.md>), [valkey](<https://devfeed.tech/topics/valkey.md>)

Tags: [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [coding-agent](<https://devfeed.tech/tags/coding-agent.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [context](<https://devfeed.tech/tags/context.md>), [pingcap-tidb](<https://devfeed.tech/tags/pingcap-tidb.md>), [real-time-analytics](<https://devfeed.tech/tags/real-time-analytics.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [what-is](<https://devfeed.tech/tags/what-is.md>)

### AI overview

Persistent AI agent memory is a durable external store that lets agents retain useful context across sessions, tools, and machines. The article distinguishes episodic, semantic, and procedural memory, and describes a capture, retrieval, and update loop with scoped access and stale-memory handling. It argues that production systems need structured metadata, tenant isolation, and consistency in addition to vector search, then presents TiDB Cloud as a database layer for agent state.

### Source excerpt

Persistent AI Agent Memory Explained: How Agents Remember Across Sessions Key Takeaways A coding agent that forgets your architecture decisions every time you open a new terminal is not broken. It is working as designed. Persistent AI agent memory is the external layer that stores and retrieves useful context across sessions, tools, and machines, so [...] The post Persistent AI Agent Memory Explained: How Agents Remember Across Sessions appeared first on TiDB.

## Why Open Source Still Matters When Agents Write the Code

DevFeed: [Why Open Source Still Matters When Agents Write the Code](<https://devfeed.tech/articles/why-open-source-still-matters-when-agents-write-the-code-65424.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/open-source-agents-write-code/>)

Author: Ed Huang

Published: 2026-10-06T15:45:08Z

Content type: opinion

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Open Source](<https://devfeed.tech/topics/open-source.md>), [open source maintenance](<https://devfeed.tech/topics/open-source-maintenance.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [maintainers](<https://devfeed.tech/tags/maintainers.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>)

### AI overview

The article argues that coding agents have made contributions cheap, shifting open source governance toward maintainers' merge decisions and encouraging private forks that reduce feedback to public projects. It contends that open source remains essential for ownership and shared infrastructure, while companies can charge for ongoing operations and accountability. The author also discusses how agents may change software markets and identifies uncertainties in that outlook.

### Source excerpt

Maintainers of well-known open source projects are dealing with a new kind of inbox. Pull requests still arrive, often in higher volume than ever, and many are clean, well-formatted, and plausible on first read. What changed is that one person can now produce a hundred of them in an afternoon, and the maintainer still has [...] The post Why Open Source Still Matters When Agents Write the Code appeared first on TiDB.

## The Database Is the Product: What Breaks When Memory Devices Scale

DevFeed: [The Database Is the Product: What Breaks When Memory Devices Scale](<https://devfeed.tech/articles/the-database-is-the-product-what-breaks-when-memory-devices-scale-64241.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/ai-hardware-data-architecture-at-scale/>)

Author: Ed Huang

Published: 2026-10-01T20:06:50Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [database-consistency](<https://devfeed.tech/topics/database-consistency.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [Boilerplate](<https://devfeed.tech/topics/boilerplate.md>), [RIOT](<https://devfeed.tech/topics/riot-os.md>)

Tags: [ai-hardware](<https://devfeed.tech/tags/ai-hardware.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [data-architecture](<https://devfeed.tech/tags/data-architecture.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [mysql-scaling](<https://devfeed.tech/tags/mysql-scaling.md>), [online-ddl](<https://devfeed.tech/tags/online-ddl.md>), [scale](<https://devfeed.tech/tags/scale.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>), [tidb](<https://devfeed.tech/tags/tidb.md>)

### AI overview

The article examines how Plaud's AI note-taking product outgrew an architecture that stored metadata in MySQL and audio and transcript files in S3. Their separate failure and transaction boundaries created consistency and retrieval concerns, while MySQL schema changes at roughly 300 million rows required maintenance windows. The team moved the metadata layer to TiDB, restoring online schema changes and reporting about 10x QPS headroom with P95 latency under 10 ms. The article argues that data architecture directly shapes product reliability and release cadence for AI memory products.

### Source excerpt

Editor's note: This post originally appeared on The New Stack and is republished with permission. The original version is available here. Imagine you just finished a two-hour meeting. You were wearing a small AI work companion that promised to capture the conversation, structure it, and let you ask questions about it later. Two hours after the [...] The post The Database Is the Product: What Breaks When Memory Devices Scale appeared first on TiDB.

## Vector and Full-Text Search Explained: Why AI Agents Need Both

DevFeed: [Vector and Full-Text Search Explained: Why AI Agents Need Both](<https://devfeed.tech/articles/vector-and-full-text-search-explained-why-ai-agents-need-both-62157.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/vector-and-full-text-search-for-ai-agents/>)

Author: Akshata Hire

Published: 2026-09-30T00:03:48Z

Content type: tutorial

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [hybrid-search](<https://devfeed.tech/topics/hybrid-search.md>), [AI-driven information retrieval](<https://devfeed.tech/topics/ai-driven-information-retrieval.md>), [elasticsearch](<https://devfeed.tech/topics/elasticsearch.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [bm25](<https://devfeed.tech/tags/bm25.md>), [full-text-search](<https://devfeed.tech/tags/full-text-search.md>), [reciprocal-rank-fusion](<https://devfeed.tech/tags/reciprocal-rank-fusion.md>), [search](<https://devfeed.tech/tags/search.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>)

### AI overview

Hybrid retrieval combines full-text keyword search and vector search to improve AI agent results: keyword matching finds exact identifiers, while embeddings find paraphrased queries. Reciprocal rank fusion merges their ranked results. Reliable agent retrieval also depends on tenant filters, permissions, and fresh transactional data; the article says Dify.AI cut infrastructure costs 80% by consolidating on TiDB Cloud.

### Source excerpt

Key Takeaways Ask an AI support agent about error E-4012 and it may come back with a confident answer about E-4021. The embeddings for the two codes sit almost on top of each other, so the agent pulled the wrong runbook, and nothing in the logs says retrieval failed. Vector and full-text search are two [...] The post Vector and Full-Text Search Explained: Why AI Agents Need Both appeared first on TiDB.

## How Plaid Migrated from Aurora MySQL to TiDB for Zero-Downtime Upgrades

DevFeed: [How Plaid Migrated from Aurora MySQL to TiDB for Zero-Downtime Upgrades](<https://devfeed.tech/articles/the-fintech-scalability-crisis-how-distributed-sql-unlocks-innovation-with-zero-downtime-operations-60338.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/fintech-scalability-crisis-how-distributed-sql-unlocks-innovation-zero-downtime/>)

Author: M. Nivanya

Published: 2026-09-26T08:31:01Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Scalability](<https://devfeed.tech/topics/scalability.md>), [Database](<https://devfeed.tech/topics/database.md>), [TiDB](<https://devfeed.tech/topics/tidb.md>), [like](<https://devfeed.tech/topics/like.md>), [migration](<https://devfeed.tech/topics/migration.md>), [Amazon Aurora](<https://devfeed.tech/topics/amazon-aurora.md>), [sharding](<https://devfeed.tech/topics/sharding.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [MySQL](<https://devfeed.tech/topics/mysql.md>), [Raft](<https://devfeed.tech/topics/raft.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [SQL](<https://devfeed.tech/topics/sql.md>)

Tags: [amazon-aurora](<https://devfeed.tech/tags/amazon-aurora.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [consensus](<https://devfeed.tech/tags/consensus.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [database](<https://devfeed.tech/tags/database.md>), [database-architecture](<https://devfeed.tech/tags/database-architecture.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [downtime](<https://devfeed.tech/tags/downtime.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [financial-services](<https://devfeed.tech/tags/financial-services.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [migration](<https://devfeed.tech/tags/migration.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [real-time-analytics](<https://devfeed.tech/tags/real-time-analytics.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [solution](<https://devfeed.tech/tags/solution.md>), [sql](<https://devfeed.tech/tags/sql.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tidb-cloud](<https://devfeed.tech/tags/tidb-cloud.md>)

### AI overview

This article examines database scalability challenges in financial technology and describes how Plaid moved from Amazon Aurora MySQL to TiDB. It presents distributed SQL as a way to scale transaction and analytical workloads while maintaining SQL compatibility, consistency, and uptime.

### Source excerpt

When Plaid's Amazon Aurora MySQL fleet faced a major-version upgrade, the estimate came back at six engineering months and tens of minutes of downtime. For a company whose APIs sit underneath thousands of financial apps, that is not a maintenance task. It is a product freeze. Plaid spent that effort moving to distributed SQL instead, [...] The post The FinTech Scalability Crisis: How Distributed SQL Unlocks Innovation with Zero Downtime Operations appeared first on TiDB.

## Vercel and TiDB Cloud Starter: The Full-Stack Playbook for AI Apps

DevFeed: [Vercel and TiDB Cloud Starter: The Full-Stack Playbook for AI Apps](<https://devfeed.tech/articles/vercel-and-tidb-cloud-starter-the-full-stack-playbook-for-ai-apps-60206.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/build-with-tidb-cloud-starter-vercel-database/>)

Author: Brian Foster

Published: 2026-09-24T20:52:55Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [TiDB](<https://devfeed.tech/topics/tidb.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [Serverless](<https://devfeed.tech/topics/serverless.md>), [Vercel](<https://devfeed.tech/topics/vercel.md>), [Database](<https://devfeed.tech/topics/database.md>), [connection pool](<https://devfeed.tech/topics/connection-pool.md>), [Deployment](<https://devfeed.tech/topics/deployment.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Environment Variables](<https://devfeed.tech/topics/environment-variables.md>), [MySQL](<https://devfeed.tech/topics/mysql.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [connection-pool](<https://devfeed.tech/tags/connection-pool.md>), [database](<https://devfeed.tech/tags/database.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [dev](<https://devfeed.tech/tags/dev.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [environment-variables](<https://devfeed.tech/tags/environment-variables.md>), [full-stack](<https://devfeed.tech/tags/full-stack.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tidb-cloud](<https://devfeed.tech/tags/tidb-cloud.md>), [tutorial](<https://devfeed.tech/tags/tutorial.md>), [vector-search](<https://devfeed.tech/tags/vector-search.md>), [vercel](<https://devfeed.tech/tags/vercel.md>)

### AI overview

A practical guide to using TiDB Cloud Starter as the database for AI applications deployed on Vercel. It explains how serverless and edge runtimes affect database connections, how to store embeddings alongside application data, and how the Vercel Marketplace integration configures connection variables.

### Source excerpt

You have an AI app running on Vercel, or a prototype that v0.dev generated in a few minutes, and now it needs a real database. The choice is harder than it looks, because serverless functions and edge runtimes break the assumptions traditional databases are built on: long-lived TCP connections, a bounded pool, and a process [...] The post Vercel and TiDB Cloud Starter: The Full-Stack Playbook for AI Apps appeared first on TiDB.

## Lakebase, TiDB X, and the Database Architecture AI Demands

DevFeed: [Lakebase, TiDB X, and the Database Architecture AI Demands](<https://devfeed.tech/articles/lakebase-tidb-x-and-the-database-architecture-ai-demands-59590.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/separation-of-compute-and-storage-lakebase-tidb-x/>)

Author: Jinpeng Zhang

Published: 2026-09-23T15:35:42Z

Content type: comparison

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Database](<https://devfeed.tech/topics/database.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [PostgreSQL](<https://devfeed.tech/topics/postgresql.md>), [databricks](<https://devfeed.tech/topics/databricks.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [database](<https://devfeed.tech/tags/database.md>), [database-architecture](<https://devfeed.tech/tags/database-architecture.md>), [database-branching](<https://devfeed.tech/tags/database-branching.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [lakebase](<https://devfeed.tech/tags/lakebase.md>), [postgres](<https://devfeed.tech/tags/postgres.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [tidb-cloud](<https://devfeed.tech/tags/tidb-cloud.md>), [tidb-x](<https://devfeed.tech/tags/tidb-x.md>)

### AI overview

This comparison examines how Databricks Lakebase and TiDB X separate durable storage from compute for database workloads involving AI coding agents, transactions, branching, and analytical access. It also describes TiDB Cloud Lake as a complementary analytical service and notes production questions that remain unresolved.

### Source excerpt

An AI coding agent can generate an application, change its schema, test several implementations, and discard most of them within a single session. Another agent might spend that session updating orders and querying months of operational history. Both need durable state, correct transactions, and fast access. They also need something query throughput doesn't measure: databases [...] The post Lakebase, TiDB X, and the Database Architecture AI Demands appeared first on TiDB.

## AI Agent State Explained

DevFeed: [AI Agent State Explained](<https://devfeed.tech/articles/ai-agent-state-explained-57650.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/ai-agent-state/>)

Author: Akshata Hire

Published: 2026-09-21T21:35:41Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [context window](<https://devfeed.tech/topics/context-window.md>), [Database](<https://devfeed.tech/topics/database.md>), [data](<https://devfeed.tech/topics/data.md>), [LangChain](<https://devfeed.tech/topics/langchain.md>), [long-context](<https://devfeed.tech/topics/long-context.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agent-state](<https://devfeed.tech/tags/agent-state.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [langchain](<https://devfeed.tech/tags/langchain.md>), [latency](<https://devfeed.tech/tags/latency.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [what-is](<https://devfeed.tech/tags/what-is.md>)

### AI overview

This article explains AI agent state as the durable record of memory, task progress, tool outputs, files, and supporting data that allows agents to continue across steps and sessions. It distinguishes persistent state from prompts, chat transcripts, and context windows, and discusses reliability needs such as recovery, atomic writes, isolation, and auditability.

### Source excerpt

Key Takeaways An agent that forgets everything between calls can't finish a multi-step task, can't recover from a mid-workflow crash, and can't show you what it actually did. That's the practical cost of statelessness, and it's the problem AI agent state exists to solve. AI agent state is the durable record of memory, task progress, [...] The post AI Agent State Explained appeared first on TiDB.

## AI agent memory requires more than persistence: selection, decay, and contamination prevention

DevFeed: [AI agent memory requires more than persistence: selection, decay, and contamination prevention](<https://devfeed.tech/articles/why-your-ai-agent-doesn-t-actually-remember-anything-55202.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/long-term-memory-ai-agents/>)

Author: Ed Huang

Published: 2026-09-17T16:14:23Z

Content type: opinion

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Persistence](<https://devfeed.tech/topics/persistence.md>), [reliability-engineering](<https://devfeed.tech/topics/reliability-engineering.md>), [Database](<https://devfeed.tech/topics/database.md>)

Tags: [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agent-reliability](<https://devfeed.tech/tags/agent-reliability.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [billing](<https://devfeed.tech/tags/billing.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [idempotency](<https://devfeed.tech/tags/idempotency.md>), [memory-fragmentation](<https://devfeed.tech/tags/memory-fragmentation.md>), [persistence](<https://devfeed.tech/tags/persistence.md>), [persistent-filesystems](<https://devfeed.tech/tags/persistent-filesystems.md>), [support](<https://devfeed.tech/tags/support.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>), [tidb](<https://devfeed.tech/tags/tidb.md>)

### AI overview

The article argues that reliable AI agent memory requires more than persistent storage. It distinguishes memory from idempotency, workflow state, and transactional consistency, and highlights selection, compression, decay, confidence weighting, and contamination filtering as key challenges.

### Source excerpt

Editor's note: This post originally appeared on The New Stack and is republished with permission. The original version is available here. A few months ago, I was reviewing a customer support agent who had a strange failure pattern. A user would chat with it on Monday about a billing issue, get partway through the resolution, and [...] The post Why Your AI Agent Doesn't Actually Remember Anything appeared first on TiDB.

## AI Coding Agent Files Explained: How Persistent Workspaces Survive Resets

DevFeed: [AI Coding Agent Files Explained: How Persistent Workspaces Survive Resets](<https://devfeed.tech/articles/ai-coding-agent-files-explained-how-persistent-workspaces-survive-resets-55201.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/ai-coding-agent-files/>)

Author: Brian Foster

Published: 2026-09-14T14:57:17Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Persistence](<https://devfeed.tech/topics/persistence.md>), [Filesystems](<https://devfeed.tech/topics/filesystems.md>), [file](<https://devfeed.tech/topics/file.md>), [Repository](<https://devfeed.tech/topics/repository.md>), [test](<https://devfeed.tech/topics/test.md>), [Database](<https://devfeed.tech/topics/database.md>)

Tags: [agent-state-stack](<https://devfeed.tech/tags/agent-state-stack.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-coding-agent](<https://devfeed.tech/tags/ai-coding-agent.md>), [database](<https://devfeed.tech/tags/database.md>), [files](<https://devfeed.tech/tags/files.md>), [persistent-filesystems](<https://devfeed.tech/tags/persistent-filesystems.md>), [repo](<https://devfeed.tech/tags/repo.md>), [test](<https://devfeed.tech/tags/test.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [what-is](<https://devfeed.tech/tags/what-is.md>)

### AI overview

This article explains that AI coding agent work includes repositories, source changes, dependencies, build and test output, logs, patches, generated artifacts, and notes--not just the final diff. It describes how ephemeral sandboxes lose this work when they reset, how persistent filesystems preserve disk-based workspace artifacts across runtimes, and why databases are still needed for metadata, permissions, and workflow state.

### Source excerpt

An agent clones a repo, installs dependencies, reproduces a failing test, and gets three-quarters of the way through a fix. The sandbox hits its duration limit. The next run starts in an empty directory with the same task, the same prompt, and none of the evidence. Nothing crashed and nothing errored. The work was never [...] The post AI Coding Agent Files Explained: How Persistent Workspaces Survive Resets appeared first on TiDB.

## Best Practices for AI-Assisted App Development: Reviewing AI-Generated Code for Production

DevFeed: [Best Practices for AI-Assisted App Development: Reviewing AI-Generated Code for Production](<https://devfeed.tech/articles/best-practices-for-ai-assisted-app-development-reviewing-ai-generated-code-for-production-43547.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/best-practices-for-developing-applications-with-tidb/>)

Author: Brian Foster

Published: 2026-09-09T07:00:00Z

Content type: tutorial

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Database](<https://devfeed.tech/topics/database.md>), [software-development](<https://devfeed.tech/topics/software-development.md>), [Code](<https://devfeed.tech/topics/code.md>), [SQL](<https://devfeed.tech/topics/sql.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [ai-assisted-coding](<https://devfeed.tech/tags/ai-assisted-coding.md>), [ai-coding-assistant](<https://devfeed.tech/tags/ai-coding-assistant.md>), [app-development](<https://devfeed.tech/tags/app-development.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [code](<https://devfeed.tech/tags/code.md>), [cursor](<https://devfeed.tech/tags/cursor.md>), [database](<https://devfeed.tech/tags/database.md>), [development](<https://devfeed.tech/tags/development.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [github-copilot](<https://devfeed.tech/tags/github-copilot.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tutorial](<https://devfeed.tech/tags/tutorial.md>)

### AI overview

A review checklist for evaluating AI-generated application code, schemas, transactions, indexes, writes, and SQL before deploying it on TiDB. It explains that coding assistants often encode single-node MySQL and PostgreSQL assumptions that can cause silent correctness bugs in distributed SQL systems.

### Source excerpt

This article shows how to efficiently develop high-quality applications with TiDB, an open-source NewSQL database that supports Hybrid Transactional and Analytical Processing (HTAP) workloads and can serve as a scale-out MySQL database without manual sharding. The post Best Practices for AI-Assisted App Development: Reviewing AI-Generated Code for Production appeared first on TiDB.

## Serverless MySQL for AI Agents: Store Memory, Tool Outputs, and Searchable State in One Backend

DevFeed: [Serverless MySQL for AI Agents: Store Memory, Tool Outputs, and Searchable State in One Backend](<https://devfeed.tech/articles/serverless-mysql-for-ai-agents-store-memory-tool-outputs-and-searchable-state-in-one-backend-43550.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/mysql-ai/>)

Author: Brian Foster

Published: 2026-09-03T21:05:13Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [MySQL](<https://devfeed.tech/topics/mysql.md>), [Scalability](<https://devfeed.tech/topics/scalability.md>), [Database](<https://devfeed.tech/topics/database.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Serverless](<https://devfeed.tech/topics/serverless.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [sessions](<https://devfeed.tech/topics/sessions.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [auto-scaling](<https://devfeed.tech/tags/auto-scaling.md>), [backend](<https://devfeed.tech/tags/backend.md>), [concurrent](<https://devfeed.tech/tags/concurrent.md>), [database](<https://devfeed.tech/tags/database.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [sql](<https://devfeed.tech/tags/sql.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [vector-search](<https://devfeed.tech/tags/vector-search.md>), [what-is](<https://devfeed.tech/tags/what-is.md>)

### AI overview

The article distinguishes three meanings of MySQL AI and focuses on using MySQL-compatible infrastructure as the backend for AI applications and agents. It explains how one auto-scaling backend can store agent memory, tool outputs, embeddings, and persistent application state while supporting SQL and vector search.

### Source excerpt

MySQL AI, in the sense that matters most to application teams, means MySQL-compatible infrastructure that holds agent memory, tool outputs, embeddings, and persistent application state in one backend, with vector search and SQL available in the same query. Serverless MySQL for AI agents is that backend delivered as an auto-scaling service rather than a cluster [...] The post Serverless MySQL for AI Agents: Store Memory, Tool Outputs, and Searchable State in One Backend appeared first on TiDB.

## What is a Serverless Database and Why It Matters for Modern AI Apps

DevFeed: [What is a Serverless Database and Why It Matters for Modern AI Apps](<https://devfeed.tech/articles/what-is-a-serverless-database-and-why-it-matters-for-modern-ai-apps-43553.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/serverless-database-2/>)

Author: Brian Foster

Published: 2026-09-01T20:10:00Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Database](<https://devfeed.tech/topics/database.md>), [Serverless](<https://devfeed.tech/topics/serverless.md>), [Scalability](<https://devfeed.tech/topics/scalability.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [systems](<https://devfeed.tech/topics/systems.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [managed-services](<https://devfeed.tech/tags/managed-services.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [serverless-database](<https://devfeed.tech/tags/serverless-database.md>), [stateless](<https://devfeed.tech/tags/stateless.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [what-is](<https://devfeed.tech/tags/what-is.md>)

### AI overview

This article explains serverless databases as managed cloud databases that separate compute from storage, automatically scale with demand, and charge based on consumption. It describes their fit for bursty AI workloads and discusses tradeoffs including cold starts, cost control, lock-in, and workloads better suited to reserved capacity.

### Source excerpt

A serverless database is a cloud database that decouples compute from storage, scales capacity automatically as demand changes, and bills for what a workload actually consumes. Servers still exist. The difference is that you never size them, patch them, or pay for idle ones, and capacity planning stops being a launch-day decision. That difference matters [...] The post What is a Serverless Database and Why It Matters for Modern AI Apps appeared first on TiDB.

## Prisma ORM with TiDB: The Serverless Setup Guide for AI Apps

DevFeed: [Prisma ORM with TiDB: The Serverless Setup Guide for AI Apps](<https://devfeed.tech/articles/prisma-orm-with-tidb-the-serverless-setup-guide-for-ai-apps-43548.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/integrating-tidb-cloud-serverless-driver-prisma-orm/>)

Author: Ramnath Sagar

Published: 2026-08-28T16:44:33Z

Content type: tutorial

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Object-relational mapping](<https://devfeed.tech/topics/orm.md>), [Prisma](<https://devfeed.tech/topics/prisma.md>), [MySQL](<https://devfeed.tech/topics/mysql.md>), [Serverless](<https://devfeed.tech/topics/serverless.md>), [Database](<https://devfeed.tech/topics/database.md>), [TypeScript](<https://devfeed.tech/topics/typescript.md>), [Node.js](<https://devfeed.tech/topics/node-js.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Cloudflare Workers](<https://devfeed.tech/topics/cloudflare-workers.md>)

Tags: [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [ai-apps](<https://devfeed.tech/tags/ai-apps.md>), [cloudflare-workers](<https://devfeed.tech/tags/cloudflare-workers.md>), [database](<https://devfeed.tech/tags/database.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [node-js](<https://devfeed.tech/tags/node-js.md>), [orm](<https://devfeed.tech/tags/orm.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [prisma](<https://devfeed.tech/tags/prisma.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tutorial](<https://devfeed.tech/tags/tutorial.md>), [typescript](<https://devfeed.tech/tags/typescript.md>)

### AI overview

A tutorial on connecting Prisma ORM to TiDB Cloud for serverless and edge applications. It explains using an HTTP driver for queries, retaining TCP connections for migrations and introspection, configuring Prisma with TiDB's MySQL compatibility, and modeling data for AI applications.

### Source excerpt

Key Takeaways Introduction Prisma is the ORM (Object-Relational Mapper) most teams use for a TypeScript project, and it connects to TiDB through the standard MySQL provider, so a schema written for MySQL works unchanged. The connection is where the work is. Serverless functions and edge runtimes open and discard connections on every invocation, which is [...] The post Prisma ORM with TiDB: The Serverless Setup Guide for AI Apps appeared first on TiDB.

## Open Weight Models Are Chapter One. The Data Layer Is the Rest of the Book.

DevFeed: [Open Weight Models Are Chapter One. The Data Layer Is the Rest of the Book.](<https://devfeed.tech/articles/open-weight-models-are-chapter-one-the-data-layer-is-the-rest-of-the-book-43552.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/open-weight-models-ai-data-layer/>)

Author: Ramnath Sagar

Published: 2026-08-24T20:29:23Z

Content type: opinion

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [AI Models](<https://devfeed.tech/topics/ai-models.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [data](<https://devfeed.tech/topics/data.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Database](<https://devfeed.tech/topics/database.md>)

Tags: [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [data](<https://devfeed.tech/tags/data.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [layers](<https://devfeed.tech/tags/layers.md>), [models](<https://devfeed.tech/tags/models.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [open-weight-models](<https://devfeed.tech/tags/open-weight-models.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>), [tidb](<https://devfeed.tech/tags/tidb.md>)

### AI overview

An opinion article argues that open-weight AI models are only the first stage of a broader sequence. It identifies memory, context, and operational data as the durable layer beneath increasingly commoditized models, drawing parallels with the adoption of open-source software and infrastructure.

### Source excerpt

On July 24, Jensen Huang made the first post of his life on X. It wasn't a product launch or a victory lap. It was a policy letter signed by 25 companies, doubling to 50 within a day, asking Washington not to restrict open weight AI models. Microsoft, Meta, IBM, a16z, and Hugging Face. The [...] The post Open Weight Models Are Chapter One. The Data Layer Is the Rest of the Book. appeared first on TiDB.

## My Journey from Traditional Monolithic Architecture to Distributed SQL

DevFeed: [My Journey from Traditional Monolithic Architecture to Distributed SQL](<https://devfeed.tech/articles/my-journey-from-traditional-monolithic-architecture-to-distributed-sql-43549.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/journey-from-traditional-monolithic-architecture-to-distributed-sql/>)

Author: Bill Kendall

Published: 2026-08-19T18:57:00Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Database](<https://devfeed.tech/topics/database.md>), [monolithic-architecture](<https://devfeed.tech/topics/monolithic-architecture.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [sharding](<https://devfeed.tech/topics/sharding.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [cloud-infrastructure](<https://devfeed.tech/topics/cloud-infrastructure.md>), [consistency](<https://devfeed.tech/topics/consistency.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-native](<https://devfeed.tech/tags/cloud-native.md>), [databases](<https://devfeed.tech/tags/databases.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [monolithic-architecture](<https://devfeed.tech/tags/monolithic-architecture.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [sharding](<https://devfeed.tech/tags/sharding.md>), [sql](<https://devfeed.tech/tags/sql.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>), [tidb](<https://devfeed.tech/tags/tidb.md>)

### AI overview

A database engineer recounts moving from vertically scaled monolithic Oracle systems to distributed SQL after cloud migrations exposed the limits of single-machine architectures. The article explains how distributed databases scale across nodes while preserving transactions, joins, relational modeling, and consistency, and connects these requirements to durable state for AI agents.

### Source excerpt

As a database veteran, I found traditional monolithic databases (such as Oracle) to have bottlenecks for mission critical applications in the cloud, which called for a next generation cloud native database to solve these challenges. This is what excited me about joining the PingCAP team, where we took the core concept of Distributed SQL's multi-node auto-sharding architecture and extended it to allow for greater flexibility and functionality beyond Spanner and other Distributed SQL databases. It is also open source and has helped to change the landscape for developers. The post My Journey from Traditional Monolithic Architecture to Distributed SQL appeared first on TiDB.

## Why Agentic AI Architecture Needs a Database, Not Just a Vector Store

DevFeed: [Why Agentic AI Architecture Needs a Database, Not Just a Vector Store](<https://devfeed.tech/articles/why-agentic-ai-architecture-needs-a-database-not-just-a-vector-store-43546.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/agentic-ai-architecture/>)

Author: Brian Foster

Published: 2026-08-10T13:00:00Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Data Infrastructure](<https://devfeed.tech/topics/data-infrastructure.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Database](<https://devfeed.tech/topics/database.md>), [Orchestration](<https://devfeed.tech/topics/orchestration.md>)

Tags: [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-ai-architecture](<https://devfeed.tech/tags/agentic-ai-architecture.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [orchestration](<https://devfeed.tech/tags/orchestration.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [vector](<https://devfeed.tech/tags/vector.md>), [vector-stores](<https://devfeed.tech/tags/vector-stores.md>), [what-is](<https://devfeed.tech/tags/what-is.md>)

### AI overview

This article explains agentic AI architecture as a system spanning models, orchestration, memory, tools, state, guardrails, and data infrastructure. It argues that production agents need authoritative records, durable workflow state, retrieval infrastructure, and auditability rather than relying only on vector stores.

### Source excerpt

Agentic AI architecture is the system design that lets an AI agent perceive context, reason over it, call tools, maintain memory, and take actions across multiple steps. It spans the model, the orchestration layer, and the data infrastructure that makes an agent's work durable rather than disposable. Most teams get the model and framework right [...] The post Why Agentic AI Architecture Needs a Database, Not Just a Vector Store appeared first on TiDB.

## Serverless Database Persistence for AI Applications: Memory, Files, Operational State, and Retrieval

DevFeed: [Serverless Database Persistence for AI Applications: Memory, Files, Operational State, and Retrieval](<https://devfeed.tech/articles/the-complete-agent-state-stack-memory-files-and-serverless-database-persistence-for-ai-apps-43554.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/serverless-database/>)

Author: Fan Wang

Published: 2026-08-07T13:00:00Z

Content type: tutorial

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Serverless](<https://devfeed.tech/topics/serverless.md>), [Persistence](<https://devfeed.tech/topics/persistence.md>), [AI Development](<https://devfeed.tech/topics/ai-development.md>), [cloud-infrastructure](<https://devfeed.tech/topics/cloud-infrastructure.md>), [Scalability](<https://devfeed.tech/topics/scalability.md>), [Database](<https://devfeed.tech/topics/database.md>), [stateless](<https://devfeed.tech/topics/stateless.md>), [event driven](<https://devfeed.tech/topics/event-driven.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agent-state](<https://devfeed.tech/tags/agent-state.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-apps](<https://devfeed.tech/tags/ai-apps.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [data](<https://devfeed.tech/tags/data.md>), [database-architecture](<https://devfeed.tech/tags/database-architecture.md>), [event-driven](<https://devfeed.tech/tags/event-driven.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [operational](<https://devfeed.tech/tags/operational.md>), [persistence](<https://devfeed.tech/tags/persistence.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [serverless-database](<https://devfeed.tech/tags/serverless-database.md>), [state](<https://devfeed.tech/tags/state.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [what-is](<https://devfeed.tech/tags/what-is.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

This guide explains how serverless databases provide managed, usage-based infrastructure that can scale to zero and support bursty AI application workloads. It examines persistence across agent memory, files, operational state, and retrieval data, and discusses evaluating connection behavior, cold starts, and system boundaries.

### Source excerpt

A serverless database is a fully-managed database that automatically scales compute and storage with demand, requires no server provisioning or capacity planning, and bills only for actual usage, including scaling to zero when idle. The provider handles infrastructure; you handle data. That definition covers the category. It doesn't cover the problem most teams actually face [...] The post The Complete Agent State Stack: Memory, Files, and Serverless Database Persistence for AI Apps appeared first on TiDB.

## Why Attend TiDB SCaiLE 2026: Same Complexity, Different Clock Speeds

DevFeed: [Why Attend TiDB SCaiLE 2026: Same Complexity, Different Clock Speeds](<https://devfeed.tech/articles/why-attend-tidb-scaile-2026-same-complexity-different-clock-speeds-43555.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/why-attend-tidb-scaile-2026/>)

Author: Brian Foster

Published: 2026-08-06T20:04:58Z

Content type: article

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Database](<https://devfeed.tech/topics/database.md>), [PostgreSQL](<https://devfeed.tech/topics/postgresql.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [systems](<https://devfeed.tech/topics/systems.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [conference](<https://devfeed.tech/tags/conference.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [event](<https://devfeed.tech/tags/event.md>), [events](<https://devfeed.tech/tags/events.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [tidb](<https://devfeed.tech/tags/tidb.md>), [tidb-cloud](<https://devfeed.tech/tags/tidb-cloud.md>), [tidb-scaile](<https://devfeed.tech/tags/tidb-scaile.md>)

### AI overview

This article explains the structure and audience of TiDB SCaiLE 2026, a one-day event focused on enterprise data complexity and AI-native applications. It highlights practitioner-led sessions, production tradeoffs, and examples involving TiDB, PostgreSQL cluster consolidation, database tenants, and agent-driven workloads.

### Source excerpt

A single user action in an agentic application no longer maps to a single database query. It spawns agent instances that branch context in milliseconds, hold memory across sessions, and provision their own tenants without waiting for a human. At Manus, agents create more than 90% of new database clusters. The stacks underneath them were [...] The post Why Attend TiDB SCaiLE 2026: Same Complexity, Different Clock Speeds appeared first on TiDB.

## Why Open-Source Data Layers Preserve Control in AI Applications

DevFeed: [Why Open-Source Data Layers Preserve Control in AI Applications](<https://devfeed.tech/articles/open-source-data-layers-will-win-the-ai-future-43551.md>)

Original publisher: [Read original article](<https://www.pingcap.com/blog/open-source-data-layer-ai/>)

Author: Terry Purcell

Published: 2026-08-05T19:20:17Z

Content type: opinion

Language: en

Sources: [TiDB](<https://devfeed.tech/sources/tidb.md>)

Topics: [Open Source](<https://devfeed.tech/topics/open-source.md>), [data](<https://devfeed.tech/topics/data.md>), [Database](<https://devfeed.tech/topics/database.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [control-plane](<https://devfeed.tech/topics/control-plane.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [control-plane](<https://devfeed.tech/tags/control-plane.md>), [data](<https://devfeed.tech/tags/data.md>), [data-platform](<https://devfeed.tech/tags/data-platform.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [thought-leadership](<https://devfeed.tech/tags/thought-leadership.md>), [tidb-cloud](<https://devfeed.tech/tags/tidb-cloud.md>)

### AI overview

This opinion article argues that AI applications make the data layer the most durable part of the stack because customer context, permissions, transactions, embeddings, conversation history, workflow state, and telemetry persist while models and frameworks change. It presents open-source data layers as a way to preserve deployment flexibility, reduce dependence on vendor roadmaps, and retain the ability to inspect, run, move, or influence the technology.

### Source excerpt

For most of the cloud era, infrastructure decisions followed a familiar pattern: choose the fastest path to production, accept some proprietary dependencies, and optimize later. AI breaks that pattern. Coding agents now reshape applications in hours, while models, frameworks, and workload shapes change underneath them. That leaves the data layer as the one place a [...] The post Open Source Data Layers Will Win the AI Future appeared first on TiDB.