Moda vs Braintrust

1. Difference 1Moda checks every trace. Braintrust scores the ones you sample.Moda's detections for agent failures run on every trace from the first one, with nothing to write. In Braintrust you write or pick each scorer, attach it to an online scoring rule, and choose a sampling rate.
2. Difference 2Moda ranks failures by the users they reach. Braintrust ranks them by share of traces.Moda groups failures into problems ranked by how many users they affect, and attributes new traces within minutes. Braintrust Patterns, in public preview, is a scheduled investigation that orders findings by severity and the share of traces affected.

What moves depends on how your app sends traces to Braintrust today. Spans from OpenTelemetry instrumentation move by changing where they are exported. Calls traced with Braintrust's own SDK move to the Moda SDK. In both cases, set your session and user IDs on the Moda keys so each agent run stays one trace under the right user, then import the history you already have in Braintrust.

1. 1.Point your OpenTelemetry exporter at Moda

   If your spans come from OpenTelemetry GenAI instrumentation, OpenLLMetry, or the Vercel AI SDK's telemetry and go to Braintrust's /otel endpoint, Moda reads them as they are. Change the exporter's endpoint to https://moda-ingest.modas.workers.dev/v1/traces, replace its headers with Authorization: Bearer and your Moda API key, and drop the x-bt-parent header. If you send spans through Braintrust's span processor, replace it with a standard OTLP exporter configured the same way. Moda does not read braintrust.* attributes such as braintrust.input or braintrust.metadata, so spans that carry their content only there need GenAI attributes or the Moda SDK.

   Moda ingest accepts uncompressed request bodies only, and the OpenTelemetry spec leaves the default compression to each SDK, so set OTEL_EXPORTER_OTLP_TRACES_COMPRESSION=none. If you export through a Collector, which compresses by default, replace the Braintrust exporter with one for Moda and set compression: none on it.

   Exporter environment variablesCopy

   export OTEL_EXPORTER_OTLP_TRACES_ENDPOINT="https://moda-ingest.modas.workers.dev/v1/traces"
   export OTEL_EXPORTER_OTLP_TRACES_HEADERS="Authorization=Bearer YOUR_MODA_API_KEY"
   export OTEL_EXPORTER_OTLP_TRACES_COMPRESSION="none"

2. 2.Replace Braintrust SDK tracing with the Moda SDK

   Braintrust's TypeScript and Python SDKs, with initLogger, wrappers such as wrapOpenAI, and the Node.js auto-instrumentation hook, log spans to Braintrust through their own logger. Their OpenTelemetry support is a separate package, so for these calls there is no exporter to repoint.

   Replace those calls with the Moda SDK for Python or Node.js. One init call instruments the OpenAI and Anthropic clients, including streamed responses and tool calls, without changes to your provider code.

3. 3.Set your session and user IDs on the Moda keys

   Moda groups spans into one trace by moda.conversation_id, or gen_ai.conversation.id, across OpenTelemetry traces, and attributes them to a user by moda.user_id. A span with neither conversation key is grouped by its own OpenTelemetry trace. Set moda.conversation_id to the session or conversation ID your app already tracks and moda.user_id to its user ID. With the Moda SDK, that is moda.conversation_id and moda.user_id in Python. In Node.js, import withContext from moda-ai and wrap each request in withContext(sessionId, userId, async () => { /* your existing agent code */ }). This isolates concurrent requests. Reserve the process-wide Moda.conversationId and Moda.userId setters for scripts handling one session at a time.

   The importer reads the user ID from user_id or userId in a trace's Braintrust metadata. Set moda.user_id to that same value, and a person's traces from before and after the switch land under one Moda user. Session IDs do not carry across the same way: the importer prefixes each imported trace ID with braintrust: and the Braintrust project ID, so a conversation that continues after the switch starts a new Moda trace.

4. 4.Import your Braintrust history

   In Settings, open Integrations, then Import traces. Choose your Braintrust region, US or EU, and paste a Braintrust API key; Moda stores it encrypted and removes it when the import completes or is cancelled. Then pick a project and a date range of up to 30 days, and run another import for each earlier window. An owner or admin starts the import, and can pause, resume, or cancel it. It runs in the background, so you can leave the page.

   Each Braintrust trace whose root span was created in the range arrives with all of its child spans: parent links, timing, inputs and outputs, models, token counts, and errors, with the original Braintrust record kept on every span. Traces whose metadata carries the same session_id, sessionId, or conversation_id become one Moda trace, and a user_id or userId becomes the Moda user. Datasets, experiments, prompts, and attachments stay in Braintrust.