← Back to Blog

Announcing Meridian Syn API v2

Today we are publicly releasing Meridian Syn API v2, the most significant update to our developer platform since launch. API v2 represents a ground-up redesign informed by eighteen months of feedback from the engineering teams building on our platform. It is faster, more consistent, more capable, and designed to scale with the increasingly sophisticated integrations our customers are building. Whether you are pulling attribution data into a custom data warehouse, triggering real-time campaign adjustments from your own systems, or building internal tools on top of Meridian Syn's intelligence layer, API v2 gives you more power with less friction.

Let me walk through the major changes. The most requested improvement was expanded endpoint coverage. API v1 covered approximately 60% of the functionality available in the Meridian Syn dashboard. If you wanted to access neural targeting scores, audience segment definitions, or experiment results, you had to either use the dashboard manually or submit a support request for a data export. API v2 provides full parity with the dashboard. Every piece of data you can see in the UI is now accessible programmatically. This includes 14 new endpoint groups: neural targeting scores and segment membership, experiment configuration and results, consent graph state and audit logs, journey analytics and path data, signal retention policies and data governance, agent fleet configuration and performance metrics, anomaly detection alerts and resolution history, custom event taxonomy and mapping, cross-device identity graph queries, real-time campaign performance streams, budget pacing and allocation recommendations, competitive signal benchmarks, predictive forecast models and scenario outputs, and webhook configuration and delivery logs. In total, API v2 exposes 87 endpoints, up from 34 in v1.

The second major improvement is the adoption of OpenAPI 3.0 as our specification standard. Every endpoint in API v2 is fully documented in an OpenAPI 3.0 spec file that is publicly accessible and versioned alongside the API itself. This means you can auto-generate client libraries in any language, import the spec directly into tools like Postman or Insomnia, and build type-safe integrations without manually reading documentation. We are also publishing official SDKs for Python, Node.js, Ruby, and Go, all generated from the OpenAPI spec and maintained by our platform team. For teams using languages we do not provide SDKs for, the OpenAPI spec makes it straightforward to generate your own. Quilmark's engineering team generated a Kotlin client from the spec in under an hour and had it deployed to production the same day.

Rate limits have been substantially improved. API v1 enforced a flat rate limit of 100 requests per minute across all endpoints, which caused problems for customers with high-throughput integrations, particularly those pulling large volumes of event-level data for warehouse synchronization. API v2 introduces tiered rate limiting. Read endpoints (GET requests) are limited to 1,000 requests per minute. Write endpoints (POST, PUT, DELETE) are limited to 200 requests per minute. Streaming endpoints, which are new in v2, have no per-request rate limit and instead throttle on data volume (10 MB per second per connection). Bulk export endpoints operate on a queue-based system with no rate limit at all, as you submit a job and receive the results asynchronously via webhook or polling. For customers on our Enterprise plan, all rate limits are configurable, and our infrastructure team can provision dedicated API capacity for particularly demanding integrations. Crestline Labs, which syncs approximately 4 million events per day from Meridian Syn into their Snowflake warehouse, reported that their sync time decreased from 47 minutes to 8 minutes after migrating to v2's bulk export endpoints.

Webhooks are the feature I am personally most excited about. API v1 was strictly request-response. If you wanted to know when something changed in Meridian Syn, you had to poll for it. API v2 introduces a full webhook system that can push real-time notifications to your systems for over 40 event types. You can subscribe to attribution model updates, experiment completions, anomaly detections, agent fleet status changes, budget threshold alerts, consent events, and more. Each webhook delivery includes a cryptographic signature for verification, automatic retry with exponential backoff for failed deliveries, and a delivery log accessible through both the API and the dashboard. We have also built a webhook testing sandbox that lets you simulate event deliveries during development without waiting for real events to occur. Vanteon's engineering team used webhooks to build a real-time alerting system that notifies their media buying team within seconds when a campaign's ROAS drops below a configured threshold, something that previously required manual dashboard monitoring.

On the authentication front, API v2 maintains backward compatibility with the API key authentication used in v1, but adds support for OAuth 2.0 with PKCE for applications that need user-scoped access. This is particularly important for partners building third-party integrations that need to access Meridian Syn on behalf of their mutual customers. We have also introduced granular API key scoping: instead of a single all-or-nothing API key, you can now generate keys with specific permission sets, for example, a key that can read attribution data but cannot modify campaign configurations. This aligns with the principle of least privilege and makes it safer to distribute API keys across different teams and systems within your organization.

Error handling has been completely overhauled. API v1 returned generic error messages that often required consulting our support team to interpret. API v2 uses structured error responses with machine-readable error codes, human-readable descriptions, and a documentation link for every error type. Validation errors include field-level detail, so you know exactly which parameter failed and why. We have catalogued over 200 specific error codes, and each one is documented with common causes and resolution steps. Our goal is that you should never need to contact support to debug an API error.

For teams currently on API v1, migration is straightforward. We have built a migration guide that maps every v1 endpoint to its v2 equivalent, with a diff of request and response schemas. The most significant breaking changes are the move from sequential numeric IDs to UUIDs for all resource identifiers, the restructuring of attribution response objects to support the new multi-model output format, and the deprecation of the legacy event ingestion endpoint in favor of the new batch event API. Our SDKs include a compatibility layer that handles the ID format change automatically. API v1 will remain operational and supported through June 2024, giving teams nine months to migrate. After that date, v1 will enter a read-only maintenance period for an additional six months before being fully decommissioned in December 2024.

We have also invested in the developer experience beyond the API itself. The new developer portal at developers.meridian-syn.io includes interactive endpoint explorers where you can make live API calls from the browser, a collection of integration recipes for common use cases (warehouse sync, real-time bidding, custom dashboards), and a community forum where developers building on our platform can share patterns and get help from both peers and our engineering team. We believe that the strength of a developer platform is not just its technical capabilities but the ecosystem around it, and API v2 is our commitment to building that ecosystem seriously.

API v2 is available today for all customers on Growth and Enterprise plans. Starter plan customers will receive access on November 1. Full documentation, SDKs, and the migration guide are live at developers.meridian-syn.io. If you have questions about the migration or want to discuss your integration architecture, our developer relations team is available at api-support@meridian-syn.io. We are excited to see what you build.

PA

Priya Anand

CTO, Meridian Syn

Your competitors are already reading this.

Request access to Meridian Syn and deploy your first autonomous agents within 48 hours. No credit card required.

Request Access