OllinAI helps engineering teams understand the risk profile of every deployment. It correlates your deployments with production incidents, scores the risk of each change with machine learning, and surfaces actionable DORA metrics across every service you run.

Platform Overview

Key capabilities

• Risk Scoring — ML-driven risk for every deployment before it ships.
• DORA Metrics — Deployment frequency, lead time, change failure rate, and MTTR.
• Incident Correlation — Automatically link incidents back to the deployments that caused them.
• ML Predictions — Forecast failure likelihood from historical signals.
• Supply Chain Security — Track provenance and surface risky dependencies.

Get your first deployment event flowing into OllinAI in four steps.

1. 1

   Sign up

   Create your free account at /sign-up. No credit card required for the trial.

2. 2

   Create an integration

   Select your CI/CD tool (GitHub Actions, GitLab CI, Jenkins, and more).

3. 3

   Add the webhook to your pipeline

   Copy the generated webhook URL and signing secret into your pipeline configuration. See Webhook Payload Format for the exact request shape.

4. 4

   Push a deployment

   Ship a change and watch it appear in the dashboard within seconds.

An integration tells OllinAI how to receive events from your CI/CD tooling. From the dashboard, open Integrations, choose your provider, and OllinAI generates a unique webhook URL and a signing secret.

• Name your integration after the pipeline or environment it represents.
• Store the signing secret securely — it is used to verify every request.
• You can create multiple integrations for different services or regions.

Once your integration exists, send a signed POST request to the webhook URL whenever a deployment completes.

curl -X POST https://api.ollinai.com/v1/events \
  -H "Content-Type: application/json" \
  -H "X-OllinAI-Signature: $SIGNATURE" \
  -H "X-OllinAI-Integration: $INTEGRATION_ID" \
  -d '{
    "commitShas": ["a1b2c3d"],
    "author": "jane@acme.com",
    "services": ["checkout-api"],
    "environment": "production",
    "deploymentTimestamp": "2026-06-07T12:00:00Z"
  }'

The dashboard tracks the four DORA metrics that correlate with high-performing engineering teams.

How to interpret trends

Each metric shows a trend arrow comparing the current period to the previous one. A change must cross a 10% threshold before a trend is shown, filtering out noise.

Insufficient data

Metrics require at least 3 deployment events before they are computed. Until then, the dashboard shows an Insufficient Data state instead of misleading numbers.

Every deployment is assigned a risk score between 0 and 1. The badge color tells you at a glance how risky a change is — see How Risk Scoring Works for the full breakdown.

The timeline plots every deployment chronologically, annotated with its risk score and any correlated incidents, so you can spot patterns across releases.

This view groups incidents alongside the deployments most likely to have caused them, ranked by temporal proximity and overlapping services.

Narrow the dashboard by service, environment, author, or time range. DORA metrics and risk distributions recompute instantly for the current filter selection.

Add a step to your workflow that posts to OllinAI after a successful deploy job.

- name: Notify OllinAI
  run: |
    curl -X POST "$OLLINAI_WEBHOOK_URL" \
      -H "Content-Type: application/json" \
      -H "X-OllinAI-Signature: ${{ secrets.OLLINAI_SIGNATURE }}" \
      -H "X-OllinAI-Integration: ${{ secrets.OLLINAI_INTEGRATION }}" \
      -d "{\"commitShas\":[\"${{ github.sha }}\"],\"author\":\"${{ github.actor }}\",\"services\":[\"checkout-api\"],\"environment\":\"production\",\"deploymentTimestamp\":\"$(date -u +%Y-%m-%dT%H:%M:%SZ)\"}"

Add a deploy-notification job to .gitlab-ci.yml that runs after your deploy stage and posts the signed payload to your webhook URL.

In your Jenkinsfile, add a post-deploy stage that uses sh 'curl ...' to send the event, reading the signing secret from Jenkins credentials.

Add a job after your deploy workflow that posts the deployment event, storing the integration ID and signature as CircleCI project environment variables.

Use a Harness Shell Script or HTTP step in your deployment pipeline to call the OllinAI webhook once the rollout succeeds.

Add a Bash or PowerShell task to your release pipeline that sends the deployment event, referencing pipeline variables for the secret and integration ID.

Configure an ArgoCD resource hook (PostSync) or notification trigger that posts to OllinAI whenever an application reaches a synced and healthy state.

Any system that can make an authenticated HTTP request can integrate with OllinAI. Send a signed POST to your webhook URL using the schema documented in Webhook Payload Format.

DORA (DevOps Research and Assessment) metrics are four research-backed measures of software delivery performance: deployment frequency, lead time for changes, change failure rate, and mean time to recovery. Taken together they balance throughput against stability.

OllinAI combines four weighted risk factors into a single normalized score for every deployment.

Risk levels

Color coding guide

Risk badges follow a consistent traffic-light scale: green for low, amber for medium, orange for high, and red for critical.

When an incident is reported, OllinAI scans recent deployments touching the affected services and ranks the most likely culprits using temporal proximity, shared services, and historical failure patterns.

OllinAI trains models on your historical deployments and incidents to predict the probability that a new change will cause a failure. As more events flow in, predictions become increasingly tailored to your team.

Tiers scale with the number of services, retention window, and access to advanced ML predictions and supply chain security features. See the Pricing page for current limits.

Every request must be signed. OllinAI verifies the X-OllinAI-Signature header (an HMAC of the raw request body using your integration's signing secret) and matches the X-OllinAI-Integration header to a known integration.

X-OllinAI-Signature:   HMAC-SHA256(body, signing_secret)
X-OllinAI-Integration: <your-integration-id>

Deployment events use the following JSON schema:

{
  commitShas: string[],
  author: string,
  services: string[],
  environment: string,
  deploymentTimestamp: string // ISO 8601
}

Required headers

• Content-Type — must be application/json.
• X-OllinAI-Signature — HMAC signature of the request body.
• X-OllinAI-Integration — your integration identifier.

Example request

curl -X POST https://api.ollinai.com/v1/events \
  -H "Content-Type: application/json" \
  -H "X-OllinAI-Signature: $SIGNATURE" \
  -H "X-OllinAI-Integration: $INTEGRATION_ID" \
  -d '{
    "commitShas": ["a1b2c3d", "e4f5g6h"],
    "author": "jane@acme.com",
    "services": ["checkout-api", "payments-worker"],
    "environment": "production",
    "deploymentTimestamp": "2026-06-07T12:00:00Z"
  }'

Response codes

• 201 — event created successfully.
• 400 — validation error in the payload.
• 401 — invalid or missing signature.

Beyond ingesting events, the REST API exposes read endpoints for deployments, incidents, and computed DORA metrics.

POST /v1/events                 Ingest a deployment event
GET  /v1/deployments            List deployments
GET  /v1/deployments/:id        Get a single deployment + risk score
GET  /v1/incidents              List incidents
GET  /v1/metrics/dora           Get computed DORA metrics

The ingestion endpoint accepts up to 100 requests per minute per integration. Read endpoints allow 600 requests per minute per account. Exceeding a limit returns 429 with a Retry-After header.

• 400 — Bad request / validation error.
• 401 — Invalid signature or unknown integration.
• 404 — Resource not found.
• 429 — Rate limit exceeded.
• 500 — Internal server error, safe to retry with backoff.