World Congress 2023 Aug 11, 2023

Trends, Challenges and Best Practices for AI at the Edge

Ekaterina Sirazitdinova

How do you deploy zero-latency AI on devices with strict compute and power limits? Learn to master quantization, pruning, and NVIDIA frameworks to conquer edge deployment.

Pause
Mute Enter Fullscreen
#1 about 7 min

Defining edge AI and its widespread industry applications

Localized data processing enables independent automation and real-time computation across manufacturing, retail, and healthcare.

#2 about 4 min

Hardware constraints and core challenges in edge environments

Operating intelligent models on embedded hardware requires balancing limited compute and power capacities with strict low-latency requirements.

#3 about 2 min

Increasing model accuracy through synthetic data generation

Augmenting realistic scenarios with synthetic data solves expensive manual labeling and effectively covers anomalous edge cases.

#4 about 7 min

Optimizing edge model throughput to ensure real-time performance

Techniques like precision quantization, network pruning, and graph optimization reduce model footprint and accelerate execution on constrained hardware.

#5 about 5 min

Accelerating AI development with software and pretrained models

Utilizing managed toolkits and foundational templates simplifies custom model training, synthetic data generation, and complex multi-node orchestration.

#6 about 6 min

Deploying optimized models for high-throughput video analytics

Specialized deployment SDKs handle layer fusion, memory management, and concurrent sequence execution for complex multi-model pipelines.

#7 about 2 min

Exploring domain gaps and hardware requirements for deployment

Correcting for texturing differences between synthetic and real datasets ensures higher accuracy across various processing units.

Matching moments

2:18 min

The necessity of Edge AI and its driving constraints

Sasha Denisov Sasha Denisov +1 · World Congress 2026 Europe

1:56 min

Optimizing AI processing capabilities for edge microcontrollers

Stephan Gillich Stephan Gillich +3 · World Congress 2024

3:21 min

Reducing cloud dependency with on-device edge AI models

Precious Osaro Precious Osaro · World Congress 2026 Europe

2:26 min

Cost and latency pressures pushing AI to the edge

Moe Sani Moe Sani · World Congress 2026 Europe

1:09 min

Deploying algorithms and AI models to edge production

Irina Terekhova Irina Terekhova · World Congress 2026 Europe

3:19 min

Technological shifts enabling practical edge AI deployment

Sasha Denisov Sasha Denisov · World Congress 2026 Europe

Upcoming sessions on this topic

Open session

World Congress 2026 North America

September 24, 2026 · 14:10–14:40

Stage 5

Edge AI: Running Agentic Intelligence Where Internet Can't Reach

Nitin Eusebius

AWS - Principal Solutions Architect

Nitin Eusebius
Open session

World Congress 2026 North America

September 24, 2026 · 14:10–14:40

Stage 7

Designing High-Performance AI APIs: Lessons from Serving Millions of Real-Time Requests

Wayne Liu

Chief Growth Officer and Americas President of Perfect Corp.

Wayne Liu
Open session

World Congress 2026 North America

September 24, 2026 · 14:10–14:40

Stage 1

Anatomy of an AI Request: Where Latency and Cost Are Really Born

Dan Fu

VP of Kernels at Together AI

Dan Fu
Open session

World Congress 2026 North America

September 25, 2026 · 15:30–16:00

Stage 7

Trust, But Verify: Continuous GPU Validation at Scale

Kyle Bell

VP of AI at TensorWave

Kyle Bell
Open session

World Congress 2026 North America

September 24, 2026 · 11:00–11:30

Tech Leaders Stage

Building AI for the Physical World

Alex Spinelli, Eystein Stenberg, Gerardo Pardo-Castellote

Alex Spinelli
Eystein Stenberg
Gerardo Pardo-Castellote
Open session

World Congress 2026 North America

September 24, 2026 · 14:10–14:40

Stage 6

Compute for your AI model: GPUs, LPUs, TPUs and beyond..

Kushaagra Goyal

Tech Lead at Rubrik, ex-CTO at Gan.AI, ex-Databricks

Kushaagra Goyal