deepstream-dev

Automate NVIDIA DeepStream pipeline construction using the pyservicemaker API.

Updated Apr 16, 2026

Search Tags:#pipeline #inference #video-analytics #gstreamer #deepstream #pyservicemaker #dynamic-sources

Ready to run

npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill deepstream-dev-sayalinvidia

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: deepstream-dev
Source: https://github.com/sayalinvidia/sayali-skills-test/tree/main/skills/deepstream-dev
Command: npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill deepstream-dev-sayalinvidia

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

NVIDIA DeepStream workflow development often requires verbose GStreamer boilerplate and intricate pipeline orchestration. This Skill provides a higher-level, Pythonic path to assemble, govern, and deploy DeepStream pipelines using the official pyservicemaker APIs, accelerating development and governance.

Core Features & Use Cases

Supports both Pipeline API (fine-grained control) and Flow API (declarative pipelines) for a broad range of use cases.
Enables dynamic source management, multi-stream batching, and integration with inference, tracking, and messaging components.
Serves as a knowledge hub with references, best practices, and concrete patterns to accelerate production-grade DeepStream deployments.

Quick Start

Create a minimal DeepStream pipeline using the Pipeline API to read a local video, perform primary inference with nvinfer, and render results on screen.

Frequently Asked Questions about deepstream-dev

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I build an NVIDIA DeepStream pipeline without writing verbose GStreamer code?▼

You can build an NVIDIA DeepStream pipeline using the pyservicemaker API, which automates pipeline construction and replaces verbose GStreamer boilerplate with a higher-level, Pythonic interface for real-time video analytics workflows.

Can I dynamically add or remove RTSP streams from a DeepStream pipeline at runtime?▼

Yes, dynamic source management is supported through REST-managed sources. This allows you to dynamically add or remove RTSP feeds and local videos from a running DeepStream pipeline without restarting the application.

What is the difference between the Pipeline API and Flow API in pyservicemaker?▼

The Pipeline API provides fine-grained control over DeepStream pipeline assembly, while the Flow API enables declarative pipeline construction. Both APIs support multi-stream batching, inference, tracking, and messaging components for video analytics.

How do I configure nvinfer for primary inference in a DeepStream application?▼

You can configure nvinfer for primary inference using YAML or INI configuration files. The pyservicemaker API assembles the DeepStream pipeline to read these configs, performing inference on local videos or RTSP feeds with optional tracking and messaging.

Does pyservicemaker support multi-stream batching and telemetry probes for DeepStream?▼

Yes, pyservicemaker supports multi-stream batching across single or multi-stream inputs and includes telemetry probes for observability. This enables production-grade DeepStream deployments with real-time video analytics and performance monitoring.

When should I use pyservicemaker instead of manual GStreamer pipeline orchestration for DeepStream?▼

Use pyservicemaker when you need to accelerate DeepStream development and governance, avoid intricate GStreamer pipeline orchestration boilerplate, and require features like dynamic REST-managed sources, multi-stream batching, and telemetry probes.