Using our SQL repo plug-in, you can collaboratively manage, save, and share your SQL code, without leaving your IDE.

AI + SQL: The New Era

SQL hasn’t changed, but the way we work with it is.

We’re entering a time where AI can boost how we write, understand, and reuse SQL, but only if it actually knows our context.

That’s the hard part. Most tools give you generic answers. Sherloq doesn’t.

We built an AI that:

  • Understands your saved queries
  • Knows how your team joins, filters, and names things
  • Sits inside your existing workflow

So instead of starting from scratch, your AI starts from what you’ve already built.

It’s not about replacing analysts.
It’s about giving them a power-up.

How it works

Sherloq builds a smart SQL layer from your actual work:

  • Every saved query becomes part of your team’s shared context
  • You get structured history, versioning, and metadata
  • The AI assistant learns from your style - filters, naming, logic

No generic answers. Just better queries based on your real work.

Ask. Fix. Explain. Modify.

Chat with your queries. Sherloq AI helps you:

  • Generate CTEs, subqueries, and more
  • Fix syntax issues in one click
  • Modify existing logic
  • Write joins and filters with the right keys
  • Understand what a query actually does

No setups. No integrations.

Sherloq lives inside your current tools:

  • Chrome editors (Bigquery, Snowflake, Databricks..)
  • DataGrip and JetBrains IDEs
  • VScode and Cursor

Install the plugin, and you’re in. No access to your data needed.
See all supported editors here.

Your SQL Knowledge Base

Sherloq automatically saves and organizes every query.
Search across versions, tag by topic, and reuse without hunting through Slack, Docs, or your memory.

Your team’s SQL brain, finally in one place, so NO MORE:

  • Trying to remember when was the last time I ran a SQL on some ‘users_v2_old’ legacy table, and who might know where I put it
  • Asking the company-wide ‘data-champs’ Slack channel if someone can send me the most updated CASE WHEN condition we talked about in last week’s guild meeting.
  • Going over 252 queries in my team’s ‘All Queries’ Docs file with cmd+f: ‘prospect_id’.

That means that it’s organized and managed, but not too stiff. It’s sits on top of our IDE, but doesn’t actually require any integration or access to the data.

It’s collaborative, but also has a place for our own personal work. It can be used by people that love data and understand it, and by those that don’t.

To get started, check out these first steps:

Need help with onboarding?

We can hop on a 15 mins quick call here or contact us to guide you through this, don’t be shy ;)