Anonymizer Service

The Anonymizer service detects personally identifiable information (PII) in text data on the NeMo Platform and replaces or rewrites it.

Overview

The service wraps the open-source NVIDIA NeMo Anonymizer library and exposes it through the NeMo Platform’s Python SDK and CLI. The library still owns PII detection, replacement, rewrite, and config validation. The platform adds inference routing through the Inference Gateway, fileset-backed inputs, plugin-service execution for streaming preview, and a Jobs-worker path for full anonymization runs.

How It Works: Library + Platform

The library defines what to anonymize and how. The platform decides where the work runs and how models are reached.

1. Build a config with the library

Use anonymizer.config (installed automatically with the nemo-anonymizer-plugin) to define the replacement strategy:

1
from anonymizer.config.anonymizer_config import AnonymizerConfig
2
from anonymizer.config.replace_strategies import Redact
3
4
config = AnonymizerConfig(
5
    replace=Redact(format_template="[REDACTED_{label}]"),
6
)

The library handles: PII detection, the four replacement strategies (Substitute, Redact, Annotate, Hash), the Rewrite mode, and config validation.

Learn more: See the open-source library documentation for detailed coverage of detection, replacement strategies, and rewrite mode.

2. Execute on the platform

Submit the config to the Anonymizer service with the NeMo Platform SDK:

1
from nemo_anonymizer_plugin.app.task_config import PreviewRequest
2
from nemo_platform import NeMoPlatform
3
4
sdk = NeMoPlatform(base_url="...", workspace="default")
5
anonymizer = sdk.anonymizer
6
7
preview_result = anonymizer.preview(PreviewRequest(
8
    config=config,
9
    data={"source": "my-fileset#data/input.csv", "text_column": "biography"},
10
    model_configs=[...],
11
    num_records=10,
12
))
13
14
preview_result.dataset           # pandas DataFrame of anonymized records
15
preview_result.trace_dataset     # detection trace
16
preview_result.display_record(0) # render a record with entity highlights

For a full anonymization run, execute the job locally or submit it to the Jobs worker:

$
nemo anonymizer run run --spec-file /path/to/run-spec.yaml      # in-process
$
nemo anonymizer run submit --spec-file /path/to/run-spec.yaml   # NeMo Services job

The SDK equivalent of run submit is sdk.anonymizer.run(request), which returns an AnonymizerJobResource you can poll with wait_until_done() and pull artifacts from with download_artifacts().

The platform handles: Inference routing through the Inference Gateway, fileset-backed inputs, and authentication.

Key Differences from Standalone Library

When using Anonymizer as a NeMo Platform service:

Replacement Strategies

The library supports four replacement strategies plus a full-passage rewrite mode. The plugin exposes all of them unchanged.

See the library documentation for the configuration shape of each strategy.

What the Plugin Adds

This package is a thin wrapper around the NVIDIA NeMo Anonymizer library. It does not re-document detection, replacement, or rewrite semantics. It adds:

• A nemo anonymizer CLI with validate, preview, and run command groups.
• An sdk.anonymizer SDK accessor (AnonymizerResource, AsyncAnonymizerResource).
• A streaming anonymizer.preview function that emits preview_dataset, trace_dataset, and failed_records frames from the plugin service.
• An anonymizer.run job that writes dataset.parquet, trace.parquet, metadata.json, and optional failed_records.json. The job can execute in the local CLI process (nemo anonymizer run run) or on the NeMo Platform Jobs worker (nemo anonymizer run submit / sdk.anonymizer.run).
• Fileset input handling (fileset://<workspace>/<fileset>#<path>).
• Inference Gateway routing for model providers referenced from model_configs.

Next Steps