Skip to content
AI ⇄ Database

Azure OpenAI to DuckDB integration — real-time data sync

Keep Azure OpenAI and DuckDB in sync without custom scripts. Cut weeks of integration work, eliminate silent data drift, and give your team a single, reliable source of truth.

  • SOC 2 and 6 other compliance frameworks
  • POC with real engineers in minutes

Adopted by fast-scaling companies moving mission-critical data in real time

Case study
Migrated from MuleSoft
Case study
Migrated from Celigo
Migrated from Heroku Connect
Migrated from Matillion
Case study
Migrated from Fivetran
Case study
Migrated from Celigo
Why teams connect Azure OpenAI and DuckDB

Sync the records in DuckDB into Azure OpenAI and land its embeddings, classifications, and generated fields back on the same rows, in real time and without a pipeline to maintain.

Azure OpenAI is a read-only source: Stacksync reads its data in real time and delivers it into DuckDB, so DuckDB always reflects the current state of Azure OpenAI — without exports, scripts, or schedulers.

AI systems do not hold customers or invoices the way business apps do. What they hold is derived from your data: the vectors and metadata in a vector store, or the classifications, extracted fields, and generated text a model produces over records it was given. DuckDB is where those source records actually live. The bridge between the two is the row itself, since an item in Azure OpenAI and the record in DuckDB it describes are two halves of the same thing, and they drift the moment one is updated without the other.

Stacksync syncs Schemas, Tables, Views, External files (Parquet/CSV/JSON) in DuckDB with Vector stores, Deployments, Models, Fine-tuning jobs in Azure OpenAI in real time. Rows created or changed in DuckDB flow into Azure OpenAI so inference and embedding run on current data, and the scores, labels, and generated fields Azure OpenAI produces flow back onto the matching rows in DuckDB, mapped field by field. A change on either side appears on the other within seconds, with no extraction job or webhook plumbing to keep alive.

Because matching is by a stable identifier, every row in DuckDB stays tied to its AI-side counterpart in Azure OpenAI. Retrieval, enrichment, and generated content always resolve back to the record they came from, so there are no orphaned vectors and no labels describing a version of a row that no longer exists.

Common use cases

  • 01 Poll Batch jobs into an operational database and fire the next pipeline step when a job's status turns to completed.
  • 02 Pull per-deployment TPM/RPM usage into a warehouse for FinOps chargeback and quota-exhaustion alerting.
  • 03 Use DuckDB as a transform step: read synced Parquet exports, aggregate with SQL, and write results back to an operational database.
  • 04 Sync SaaS data to Parquet on object storage and query it with DuckDB without standing up a warehouse.

Common sync patterns

Run the AI on current data

Rows created or changed in DuckDB flow into Azure OpenAI as they happen, so embeddings, classifications, and prompts run on the latest records instead of a nightly snapshot.

Write results back onto the record

Scores, labels, extracted fields, or generated text produced in Azure OpenAI land on the matching row in DuckDB, next to the source data your applications already query.

Keep derived data fresh as sources change

When a row in DuckDB is updated or removed, its counterpart in Azure OpenAI is updated or removed too, so nothing in Azure OpenAI describes a record that has since changed or gone.

What you can sync between Azure OpenAI and DuckDB

Representative objects on each side — any object or custom field can map to any target. Schemas are auto-detected; types are converted between the two systems.

Azure OpenAI objects DuckDB objects How this pairing syncs
Usage and quota Per-deployment TPM/RPM consumption and remaining quota, read from usage endpoints and Azure Monitor for cost and throttling reporting. External files (Parquet/CSV/JSON) Files DuckDB queries in place without loading, common as a sync interchange format. Usage and quota is specific to Azure OpenAI and External files (Parquet/CSV/JSON) to DuckDB — each maps to any object or custom field on the other side.
Assistants Persistent assistants (preview) with instructions, tools, and linked files; read as configuration inventory, not authored via sync. Attached databases Additional database files or external systems attached into one session for cross-source queries. Assistants is specific to Azure OpenAI and Attached databases to DuckDB — each maps to any object or custom field on the other side.
Vector stores File collections (preview) backing file-search retrieval; read as metadata such as name, file counts, and status. Database files Single-file .duckdb databases that jobs read and write directly on disk or object storage. Vector stores is specific to Azure OpenAI and Database files to DuckDB — each maps to any object or custom field on the other side.
Deployments Named model deployments (model, version, SKU, assigned TPM capacity) read as a control-plane inventory via Azure Resource Manager; read-only in sync. Schemas Namespaces within a database used to organize tables in sync outputs. Deployments is specific to Azure OpenAI and Schemas to DuckDB — each maps to any object or custom field on the other side.
Models Catalog of base and fine-tunable models available per region; read-only reference data used to resolve deployment and fine-tuning targets. Tables Columnar tables created via SQL; the destination for materialized sync data. Models is specific to Azure OpenAI and Tables to DuckDB — each maps to any object or custom field on the other side.
Fine-tuning jobs Training jobs with status, base model, hyperparameters, and result files; status is polled from queued through succeeded or failed. Views SQL views used to shape or filter data for downstream consumers. Fine-tuning jobs is specific to Azure OpenAI and Views to DuckDB — each maps to any object or custom field on the other side.

How changes propagate between Azure OpenAI and DuckDB

Each direction of the sync is driven by what the source system can signal and what the destination accepts — detection, delivery, and expected latency below.

Azure OpenAI DuckDB Interval-based propagation

DetectionStacksync polls Azure OpenAI for changes on an incremental schedule, reading only records changed since the previous pass. Polling: list endpoints plus GET on job IDs for status.

DeliveryEach detected change is applied to DuckDB as a row-level write, with types converted between the two schemas.

DuckDB Azure OpenAI Interval-based propagation

DetectionStacksync polls DuckDB for changes on an incremental schedule, reading only records changed since the previous pass. Polling or full re-reads.

DeliveryAzure OpenAI does not accept inbound record writes, so this direction carries requests rather than records: Azure OpenAI's output flows back as field updates on the originating DuckDB records.

Rate-limit considerations

  • Azure OpenAI: Per-deployment TPM and RPM limits (about 6 RPM per 1000 TPM), scoped by region and subscription; control-plane ARM calls throttle separately.
  • DuckDB: No API rate limits; throughput is bounded by local compute and I/O.
What ships with Azure OpenAI ⇄ DuckDB

Connect Azure OpenAI and DuckDB for flexible, real-time data sync.

Real-time sync, workflow automation, event queues, EDI, and monitoring, for every Azure OpenAI–DuckDB connection.

Real-time

Real-time sync

Changes in Azure OpenAI or DuckDB instantly reflect in both systems. No stale data, no manual imports.

No-code + pro-code

Workflow automation

Trigger automated workflows whenever Azure OpenAI or DuckDB data changes, update records, fire webhooks, or kick off sequences without brittle API scripts.

At scale

Event queues

Handle millions of events per minute without losing a single Azure OpenAI or DuckDB record.

Observability

Monitoring

Track your Azure OpenAI ⇄ DuckDB sync health, view errors, and replay failed events in one click.

Trading partners

EDI

Transform legacy EDI complexity into simple database interactions between Azure OpenAI and DuckDB.

How the Azure OpenAI and DuckDB connectors work

Azure OpenAI

Integration surface
REST data-plane (inference + authoring) and Azure Resource Manager control-plane
Authentication
API key in the api-key header, or a Microsoft Entra ID bearer token / managed identity
Change detection
Polling: list endpoints plus GET on job IDs for status; no webhooks or change feed. Fine-tuning and batch jobs expose queued/running/succeeded states.
Capabilities
read
Rate limits
Per-deployment TPM and RPM limits (about 6 RPM per 1000 TPM), scoped by region and subscription; control-plane ARM calls throttle separately.

DuckDB

Integration surface
In-process SQL engine via client libraries (Python, Node.js, JDBC, CLI); no server or network API by default
Authentication
None built in; access control is file-system level (MotherDuck adds token auth for its hosted service)
Change detection
Polling or full re-reads; no change feed or transaction log API
Capabilities
read · write
Rate limits
No API rate limits; throughput is bounded by local compute and I/O
How it works

How to connect Azure OpenAI to DuckDB — three steps, no code

Configure and sync within minutes, no code. Whether you sync 50k or 100M+ records, Stacksync handles the queues, infra, and plumbing. Integrations are non-invasive and need zero setup on your systems.

  1. 01

    Connect your apps

    Authenticate Azure OpenAI and DuckDB with each platform's native method — OAuth, API keys, or service accounts — plus secure options like SSH tunneling, IP whitelisting, and VPC peering.

    • OAuth 2.0
    • SSH tunnel
    • VPC peering
    Azure OpenAI connected
    DuckDB connected
    OAuth 2.0
    SSH tunnel
    SSL certificate
    VPC peering
  2. 02

    Choose tables

    Pick the Azure OpenAI and DuckDB objects to sync — Stacksync auto-detects both schemas, including custom fields where the platform exposes them. Sync to existing tables, or let Stacksync create new ones with ideal data types.

    • Standard objects
    • Custom objects
    • Auto-schema
    objects · Azure OpenAI ⇄ DuckDB
    Customers 12,480
    Sales Orders 8,213
    Invoices 5,902
    Items 1,344
  3. 03

    Map fields

    Fields map automatically even when names and types differ. Stacksync handles transformation and type casting for you, zero configuration required.

    • Auto-map
    • Type casting
    • Transforms
    Azure OpenAI DuckDB
    Company company_name text
    Email email text
    Amount amount numeric
    Created created_at timestamp
Try interactive demo
FAQ

Azure OpenAI and DuckDB integration FAQ

SECURITY

Security teams trust Stacksync

As a data company, we understand the importance of keeping your data secure. Stacksync is built with security best practices to keep your data safe at every layer, and is DPF-certified for US, EU, UK and CH data transfers.

Learn more about security
SOC 2 Type II
ISO 27001
HIPAA BAA
GDPR
CCPA
DPF US-EU-UK-CH
→ SECURITY WITH BENEFITS

SSO & SCIM

Let your users access Stacksync from your centralized user management systems. Works with Okta, Azure, Google SSO and more.

Alerts

Immediately get alerted about record syncing issues over email, Slack, PagerDuty and WhatsApp. Resolve issues from a centralized dashboard with retry and revert options.

Secure connection options

Securely connects to your systems with:

OAuth 2 SSH Tunnelling SSL certificates IP Whitelisting VPN gateway VPC peering and more

Related integrations

Every pair below is a real-time, two-way sync. Search all 421 integrations available for Azure OpenAI and DuckDB.

Popular · 7 of 421
All Azure OpenAI integrations All DuckDB integrations Browse all integrations
Coworkers laughing in front of a laptop in a casual office setting

Your last integration took months.
Your next one takes a prompt.

Book a demo Tour the platform on your own