Ask HN:给 AI 代理数据库访问权限,会是新的商业智能工具问题吗?
2 分•作者: vira28•3 个月前
我正在为几家处于早期到中期阶段的初创公司提供咨询,他们都问了同一个问题:他们的 AI/ML 团队需要生产环境的 Postgres 数据,但没人确切知道该如何提供给他们。<p>我之前处理过类似的情况,是为 BI 团队提供数据——使用带有较大 `max_standby_streaming_delay` 和 `hot_standby_feedback` 设置的只读副本,并接受主库偶尔出现的膨胀。效果还不错。但 AI/ML 的需求感觉有所不同,但我还无法完全表达出来,这也是我提问的部分原因。<p>我正在尝试校准以下几点:<p>代理实际上连接到哪里?主库(带有 RLS),只读副本,数据仓库(Snowflake/BigQuery/Redshift),湖仓一体(S3 上的 Iceberg/Delta),还是其他地方?<p>如果你没有这样做——是出于合规性、成本顾虑、糟糕的体验(失控的查询、提示语中的 PII),还是其他原因?<p>而我最感兴趣的是:这与向 BI 工具提供数据库访问权限是否真的有所不同,还是只是换了件新衣服的同一个问题?<p>我不是在寻找产品推荐。我试图从真正面临过这个挑战的人那里获得真实的感受。
查看原文
A couple of early-to-mid-stage startups I'm consulting with are asking the same question: their AI/ML team wants production Postgres data, and nobody's quite sure how to give it to them.<p>I've handled this before for BI teams — read replica with a generous `max_standby_streaming_delay` and `hot_standby_feedback` on, accepting the occasional bloat on the primary. Worked fine. But the AI/ML ask feels different in ways I can't fully articulate yet, which is part of why I'm asking.<p>A few things I'm trying to calibrate:<p>Where does the agent actually connect? Primary with RLS, read replica, warehouse (Snowflake/BigQuery/Redshift), lakehouse (Iceberg/Delta on S3), or something else?<p>If you're <i>not</i> doing this — is it compliance, cost fear, bad experiences (runaway queries, PII in prompts), or something else?<p>And the one I'm most curious about: does this actually feel different from giving BI tools DB access, or is it the same problem wearing new clothes?<p>Not looking for product recommendations. Trying to get a real sense from people who've actually faced this challenge.