InfoQ Homepage Presentations
-
From S3 to GPU in One Copy: Rethinking Data Loading for ML Training
Onur Satici discusses Vortex, an open-source columnar file format designed to bypass CPU bottlenecks, enabling ultra-fast S3-to-GPU data streaming and dynamic query pruning at up to 60 Gbps.
/presentations/vortex-columnar-file-format-gpu-streaming/en/mediumimage/onur-satici-medium-1787813397611.jpeg)
-
Instrumentation at Scale: Having Your Performance Cake and Eating It Too
Brian Martin explains how to achieve high-performance instrumentation without massive runtime overhead, comparing Rust metrics libraries, counter designs, and eBPF system telemetry techniques.
/presentations/instrumenting-scale/en/mediumimage/brian-martin-medium-1787812453109.jpg)
-
Beyond Prompting: Context Engineering for Production-Grade AI
Ricardo Ferreira explains how context engineering overcomes LLM latency, cost, and memory limitations in AI systems, sharing architectural lessons from building a custom Alexa backend with Redis.
/presentations/context-engineering-redis-llm-architecture/en/mediumimage/ricardo-ferreira-medium-1787820318843.jpg)
-
Beyond Line Charts: Why Some Diversity in Telemetry Visualization Is Long Overdue
Yao Yue explains why line charts often fail during production incidents and shares how software architects & engineering leaders can transform telemetry data to answer critical operational questions.
/presentations/telemetry-data/en/mediumimage/yao-yue-medium-1787812332305.jpeg)
-
Running AI at the Edge: Running Real Workloads Directly in the Browser
James Hall explains why engineering teams should shift AI workloads on-device. He shares local inference strategies, WebGPU optimization techniques, and architecture choices for zero-trust privacy.
/presentations/local-ai-browser-inference-privacy/en/mediumimage/james-hall-medium-1787813225372.jpeg)
-
Architecting the Data Layer for AI Agents: from Transactional Systems to MCP and Semantic Models
Fabiane Nardon explains how to architect enterprise data platforms for AI agents, balancing precision, security, and token costs using MCP, data mesh, and semantic web technologies.
/presentations/enterprise-data-architecture-ai-agents/en/mediumimage/fabiane-nardon-medium-1787218382028.jpeg)
-
From DVDs to Global Streaming: How Netflix’s Commerce Architecture Actually Evolved
Kasia Trapszo explains how Netflix evolved its commerce platform, making pragmatic architectural trade-offs through global expansion, regulatory shifts, domain splits, and live-event scaling.
/presentations/netflix-commerce-architecture-evolution/en/mediumimage/kasia-trapszo-medium-1787217920531.jpg)
-
Python, Numba, and Algorithm Design: Building Efficient Models in Financial Services
Chad Schuster explains how using Numba JIT compilation and GPUs speeds up compute-intensive Python models, driving 10x cloud cost reductions while navigating architecture trade-offs.
/presentations/numba/en/mediumimage/chad-schuster-medium-1787217755995.jpeg)
-
Can Claude Fix Itself? Using LLMs for Incident Response
Anthropic's Alex Palcuie shares how LLMs transform incident response, highlighting where Claude excels at log analysis and why automated AI SREs still can't replace human judgment.
/presentations/claude-sre-incidents/en/mediumimage/alex-palcuie-medium-1786539336984.jpg)
-
Continuous Delivery for Foundational Platforms
Ian Nowland explains why standard CI/CD models fail for foundational platforms, sharing lessons from AWS and Datadog on how engineering leaders can implement safe progressive deployments.
/presentations/cd-reliability-innovation/en/mediumimage/ian-nowland-medium-1787217621961.jpg)
-
Prompt to Prod: Engineering an Autonomous SDLC at Scale
Andrew Swerdlow shares how Roblox transitions from AI autocomplete to fully autonomous software development with "Prompt to Prod," covering safety guardrails, infrastructure, and metrics.
/presentations/autonomous-ai-software-development-roblox/en/mediumimage/andrew-swerdlow-medium-1787218240356.jpg)
-
SafeChat: Building AI-Powered Safety Systems at Scale in a Real-Time Marketplace
Bruna Pereira shares how DoorDash built a scalable AI moderation platform. Learn how combining cheap classifiers with LLM scoring reduced incidents and cut latency in real-time chat.
/presentations/doordash-llm-ai-moderation-platform/en/mediumimage/bruna-pereira-medium-1786539036040.jpeg)
/presentations/ai-lead-interview/en/mediumimage/daniel-medium-1784808762941.jpeg)
/presentations/ai-framework-data-infrastructure/en/mediumimage/medium-1784107102579.jpeg)
/presentations/agent-software-engineering/en/mediumimage/jake-medium-1784183841742.jpeg)
/presentations/otel-slm-ai/en/mediumimage/benomahony-medium-1783500827260.jpeg)
/presentations/instrumenting-scale/en/mediumimage/brian-martin-medium-1787812453109.jpg)
/presentations/telemetry-data/en/mediumimage/yao-yue-medium-1787812332305.jpeg)
/presentations/numba/en/mediumimage/chad-schuster-medium-1787217755995.jpeg)
/presentations/cd-reliability-innovation/en/mediumimage/ian-nowland-medium-1787217621961.jpg)
/presentations/architecture-patterns-llm/en/mediumimage/JendrikJordening-medium-1786535626643.jpeg)
/presentations/emulation-genai/en/mediumimage/KennedyTorkura-medium-1785396173788.jpeg)
/presentations/agentic-compute/en/mediumimage/ArunJoseph-medium-1785311900135.jpeg)
/presentations/kotlin-serverless/en/mediumimage/elena-medium-1784808530534.jpeg)