Introduction
RavenDB, the multi‑model NoSQL document database trusted by over 12,000 customers, announced Quill, a context layer that sits on top of existing PostgreSQL, SQL Server, and MySQL databases. Quill makes it possible to launch production‑grade AI agents directly against an organization’s system of record without moving data or building a custom AI stack from scratch.- --
The Enterprise AI Integration Challenge
Many IT departments still rely on legacy SQL systems that pre‑date modern AI technologies such as embeddings and large language models (LLMs). While AI agents promise conversational access to data, integrating them with these entrenched databases is fraught with obstacles:
- Data migration risk – moving critical tables can break downstream applications.
- Complex pipeline engineering – building semantic search, retrieval, security, and governance layers often takes 18–24 months.
- Scalability concerns – proof‑of‑concept agents lack the plumbing needed for production workloads.
- --
What Is Quill?
Quill is a database‑native context layer that connects directly to an organization’s existing SQL database. It provides:
1. Semantic search and vector indexing built into the database.
2. AI‑ready APIs that expose data as embeddings for LLMs.
3. Out‑of‑the‑box agents that can be deployed to web chat, WhatsApp, Telegram, Slack, and Discord.
4. Governance, security, and audit trails that keep the source system authoritative.
“Anyone can stand up an AI demo in an afternoon, but getting that demo into production with data pipelines, semantic search, security, governance, all the plumbing a small proof of concept doesn’t need until it has to run at scale, is the hard part,” – Oren Eini, Founder & CEO, RavenDB.
- --
How Quill Works
1. Connect – Point Quill at the existing SQL instance (PostgreSQL, SQL Server, or MySQL).
2. Context Layer Creation – Quill builds a vector index of the tables you select and generates metadata needed for LLM retrieval.
3. Agent Configuration – Either let the LLM draft an agent based on your data or write the configuration manually via the RavenDB UI.
4. Deployment – Publish the agent to the desired channel (web chat, Slack, etc.). The agent runs inside the database, keeping latency low and security tight.
Quick Start Checklist
- Choose your LLM provider (OpenAI, Azure, Anthropic, etc.).
- Install the Quill connector on the target SQL server.
- Define the tables/columns to be indexed.
- Create the agent via the RavenDB console or let the LLM generate one.
- Enable the desired communication channel.
- --
Benefits and Use Cases
| Benefit | Traditional In‑House Build | Quill Approach |
|---------|----------------------------|----------------|
| Time to production | 18‑24 months | 4‑6 weeks |
| Migration required | Yes | No |
| Custom AI stack | Must be built | Provided out‑of‑the‑box |
| Ongoing maintenance | High overhead | Managed by RavenDB |
| Supported channels | Custom development | Web chat, WhatsApp, Telegram, Slack, Discord |
Real‑World Scenarios
- Customer support – Agents can answer billing queries by reading directly from the finance SQL tables.
- Sales intelligence – Pull up the latest pipeline data without exposing raw tables to the LLM.
- Compliance reporting – Generate regulatory summaries while the source system remains the single source of truth.
- --
Comparison with Other Approaches
| Approach | Data locality | Migration | Development effort | Scalability |
|----------|----------------|-----------|-------------------|------------|
| Direct SQL generation by LLMs (DIY) | On‑prem DB | Often required for schema alignment | High (prompt engineering, safety guards) | Variable |
| External vector store + API layer | Separate service | Yes (ETL) | Medium (pipeline code) | Needs extra ops |
| Quill (RavenDB) | Native inside DB | No | Low (configuration UI) | Built‑in, ACID‑safe |
- --
Future Outlook
Quill positions RavenDB as a bridge between the old world of relational data and the new world of generative AI. By delivering a production‑ready stack in weeks, it addresses the primary cause of AI project failure—complex integration—while keeping the authoritative data source untouched. Gartner’s 2025 Hype Cycle already flags AI agents as a fast‑rising technology; Quill could become the de‑facto standard for enterprises that cannot afford massive data migrations.
- --
Conclusion
RavenDB’s Quill transforms enterprise SQL databases into AI‑ready knowledge stores without the pain of migration or custom stack development. Teams can spin up agents that converse over Slack, WhatsApp, or a web UI in a matter of weeks, gaining immediate business value while preserving data integrity and governance. For organizations wrestling with legacy data and ambitious AI roadmaps, Quill offers a pragmatic, production‑grade shortcut.
- --
1. Getting started: Adding an AI agent – https://docs.ravendb.net/quill/getting-started/adding-an-ai-agent
2. RavenDB Launches Quill – https://techstartups.com/2026/09/08/ravendb-launches-quill-to-bring-production-ai-agents-to-enterprise-sql-systems-no-migration-required
3. RavenDB brings AI agents to existing SQL databases – https://www.techzine.eu/news/data-management/144146/ravendb-brings-ai-agents-to-existing-sql-databases-with-quill
4. RavenDB Launches Database‑Native AI Agent Creator – https://ravendb.net/press-releases/ravendb-launches-the-first-fully-integrated-database-native-ai-agent-creator
