Vercel
Aient is an AI SRE for your Vercel apps
Point a Vercel Drain at Aient and your function logs and traces flow straight into detection and remediation — no SDK, no code change. Aient reads the stack traces inside your Vercel errors, finds the root cause against your source, and opens a reviewed GitHub fix PR.
From a Vercel Drain to a fix PR
A Drain forwards telemetry Vercel already collects. Aient turns it into detected problems and, for crash-class issues, a reviewed pull request — without you instrumenting anything.
1
Add a Drain in Vercel
Generate a Drain URL and signing secret in Aient under Settings → Environments → Vercel Drains, then in Vercel go to Settings → Drains → Add Drain, choose Logs or Traces, and paste them into a Custom Endpoint.
2
Aient detects the problem
Vercel function errors and trace failures arrive in Aient and are grouped into scored problems — broad detection across logs and traces, no dashboards to build.
3
A reviewed fix PR lands
For crash-class problems, Aient investigates the root cause against your source code and opens a GitHub pull request. You review and merge; Aient then verifies the fix held in production.
The full step-by-step, with screenshots of Vercel's Add Drain flow, is in the Drain setup guide (opens in a new tab).
Broad detection, crash-class auto-fix
Aient detects broadly across your Vercel logs and traces. Closed-loop auto-fix runs on the crash-class subset — exceptions that resolve to a stack frame in a repo Aient can read — and you can request remediation on anything else.
Logs
Vercel function logs — lambda, edge, build and more. Error and fatal records surface as problems, and the stack traces inside them feed automated root cause analysis.
Traces
OpenTelemetry traces over OTLP/HTTP. Aient ingests them as-is, keeping the project and deployment context Vercel attaches, so distributed errors show up with the spans behind them.
Aient proposes; you dispose
Each Drain URL is unique to your organisation, and Aient verifies a signature on every request using a secret you set in Vercel. A leaked URL alone cannot be used to inject telemetry, and each organisation's data is isolated.
Remediation opens a reviewed GitHub pull request. By default the merge button stays yours; merge automation is opt-in per environment and only acts after a human approves the exact PR head and checks pass. After the merge, Aient watches production to confirm the problem is actually resolved, and reopens it if it regresses.
Pricing is the same as everywhere: EUR 99/month base plus usage for the AI work, with a budget you set. See the pricing page for how base-plus-usage billing applies.
Questions, answered
How does Aient connect to Vercel?
Through a Vercel Drain. A Drain forwards your project's logs and OpenTelemetry traces to any HTTPS endpoint. In Aient, under Settings → Environments → Vercel Drains, you generate a URL and signing secret, then paste them into a Vercel Custom Endpoint drain. There is no SDK to install and no code change — it is the near-zero instrumentation path.
What does Aient do with my Vercel logs and traces?
It detects problems. Error- and fatal-level Vercel function logs, and trace errors, are grouped into problems, the same way Aient handles SDK telemetry. For crash-class problems that resolve to a stack frame in a repo Aient can read, it runs automated root cause analysis and opens a reviewed GitHub pull request with the fix.
Will this reduce my MTTR?
That is the intent. Aient compresses the slow part of an incident — triage and writing the fix — into an automated investigation that ends in a ready-to-review pull request. The Drain removes the setup step, so a Vercel app reaches closed-loop AI incident management quickly. Final MTTR still depends on your own review and deploy.
Do I need both a Logs drain and a Traces drain?
For full coverage, yes. Vercel allows one data type per drain, so you create one Logs drain and one Traces drain, each pointed at its own Aient URL. Logs carry the error and fatal records behind most crash-class problems; traces carry distributed context.
Is it secure?
Each Drain URL is unique to your organisation, and Aient verifies a signature on every request using a secret you configure in Vercel, so a leaked URL alone cannot be used to inject telemetry. Telemetry is isolated per organisation, and remediation opens a pull request rather than merging — you stay in command.
Does Vercel ingestion cost extra?
No separate fee. It is part of Aient (EUR 99/month base plus usage). Ingesting Vercel logs and traces is included; the remediation work Aient performs is billed as usage against a budget you set, with live spend tracking.
Connect your Vercel app in minutes
Add a Drain, paste the Aient URL and secret, and let Aient detect — and fix — what's breaking in production. You review every pull request before it ships.