{"slug": "we-put-an-llm-in-a-postgresql-analyzer-then-stopped-letting-it-decide-truth", "title": "We put an LLM in a PostgreSQL analyzer, then stopped letting it decide truth", "summary": "Elevarq released Elevarq Analyzer 1.0, a PostgreSQL monitoring and investigation tool that uses deterministic analysis to establish findings from evidence before using a language model to explain them, ensuring the model does not decide what is true. The tool runs inside the user's environment and can deliver implementation-ready tickets to GitHub, GitLab, Jira, or Linear, but does not execute changes against PostgreSQL.", "body_md": "# Elevarq Analyzer 1.0: Evidence before explanation\n\nToday we're releasing Elevarq Analyzer 1.0.\n\nPostgreSQL monitoring and PostgreSQL investigation are two different jobs.\n\nMonitoring tells you that CPU increased, queries slowed down, or IO changed. Sometimes it tells you nothing while a problem develops underneath.\n\nThe next step is investigation. What changed? Does it matter? Is this normal for the workload? What evidence supports the diagnosis? And what should we do about it?\n\nAn experienced PostgreSQL DBA knows where to start looking.\n\nBut even an experienced DBA has limited time. Nobody investigates every small anomaly when there are incidents to handle, migrations to prepare, and releases to ship.\n\nThe small things wait. Most of them stay small. Some don't.\n\nElevarq Analyzer keeps doing that investigative work across every database, including the things nobody has time to investigate today.\n\n## How it works\n\nElevarq Analyzer has three parts.\n\nSignals collects read-only evidence from PostgreSQL.\n\nDeterministic analysis evaluates that evidence and establishes findings.\n\nInsight explains those findings so an engineer can understand and act on them.\n\nWhen a finding requires action, it can be delivered to GitHub, GitLab, Jira, or Linear as an implementation-ready ticket, with the evidence and recommended change included.\n\nThe workflow stops there. Elevarq Analyzer doesn't execute anything against PostgreSQL. Your team reviews the recommendation and decides. Database changes depend on context that lives outside database statistics: release schedules, application behavior, maintenance windows, business priorities.\n\n## A plausible explanation is not a finding\n\nModern language models know a lot about PostgreSQL. Give one enough statistics and context, and it will often produce an explanation that sounds like something an experienced engineer would say.\n\nThat is useful. It is also dangerous.\n\nConsider HOT updates. PostgreSQL can avoid creating new index entries when an UPDATE changes no indexed columns and the new tuple version can be stored on the same heap page. When HOT updates happen less often than expected, insufficient free space on the page is one possible cause, and lowering fillfactor can be a remedy.\n\nA PostgreSQL-aware model knows this. Show it a table with a low HOT ratio and it may tell you to lower fillfactor.\n\nThat might be right. But there is another possible cause: the UPDATEs touch an indexed column. If that's what's happening, fillfactor won't solve the problem. Unless the evidence establishes why HOT updates aren't occurring, the causes cannot be told apart.\n\nWhat the evidence establishes is that HOT behavior deserves attention. The fillfactor diagnosis is a hypothesis.\n\nDuring pre-release testing we kept finding cases like this: explanations that were technically plausible but went further than the evidence justified.\n\nA technically correct explanation can still be an incorrect finding.\n\n## No evidence, no finding\n\nTelling the model to be careful helps. It is not enough. A language model is built to produce the most useful continuation, and supplying the likely explanation is usually what makes it useful.\n\nWe don't want the most plausible explanation. We want the explanation the evidence supports.\n\nSo in Elevarq Analyzer, the language model does not decide what is true. Deterministic analysis establishes findings from the evidence. Insight explains them. If the evidence needed for a finding isn't there, the finding isn't reported.\n\nThat can make an individual report look less clever. We think it makes the product more useful in production. An engineer reading a finding should never have to wonder whether it was established from evidence or guessed by a model.\n\n## Your telemetry stays with you\n\nElevarq Analyzer runs inside your own environment. Database metadata, schema, statistics, and workload characteristics are not things most organizations want to send to an external service.\n\nAnalysis and explanation run locally. No database telemetry leaves your environment.\n\n## What 1.0 means\n\n1.0 does not mean Elevarq Analyzer knows every PostgreSQL failure mode. It doesn't. Detection will expand. Signals will collect more evidence. Models and explanations will improve.\n\n1.0 means we stand behind the methodology.\n\nFour principles shaped this release:\n\n**Deterministic detection first.** Explanation follows evidence.\n\n**No automatic changes.** Elevarq Analyzer recommends. Your team decides.\n\n**Telemetry stays in your environment.** Analysis and explanation run locally.\n\n**No evidence, no finding.** A plausible explanation isn't enough.\n\nThat last one took more work than we expected. It was worth it.\n\nElevarq Analyzer 1.0 is available now. We're onboarding the first customers directly.", "url": "https://wpnews.pro/news/we-put-an-llm-in-a-postgresql-analyzer-then-stopped-letting-it-decide-truth", "canonical_source": "https://blog.elevarq.com/elevarq-1-0", "published_at": "2026-09-08 01:15:24+00:00", "updated_at": "2026-09-08 01:31:34.229770+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-tools", "ai-products"], "entities": ["Elevarq", "Elevarq Analyzer 1.0", "PostgreSQL", "GitHub", "GitLab", "Jira", "Linear"], "alternates": {"html": "https://wpnews.pro/news/we-put-an-llm-in-a-postgresql-analyzer-then-stopped-letting-it-decide-truth", "markdown": "https://wpnews.pro/news/we-put-an-llm-in-a-postgresql-analyzer-then-stopped-letting-it-decide-truth.md", "text": "https://wpnews.pro/news/we-put-an-llm-in-a-postgresql-analyzer-then-stopped-letting-it-decide-truth.txt", "jsonld": "https://wpnews.pro/news/we-put-an-llm-in-a-postgresql-analyzer-then-stopped-letting-it-decide-truth.jsonld"}}