Beyond the Notebook: Shipping AI Agents to Production (Part 2)
詳細
Your demo works. Your POC impressed the stakeholders. Now what? Memory, grounding, safety controls, cost management, quality evaluation - the 98% of the work that demos skip. This is Part 2 of our series on building Agentic AI systems. 200+ members showed up to Part 1, voted on books for the next session, and this one came out on top.
Suhas Suresha (Adobe) and Dewang Sultania (Netflix) are bringing their Manning book "Designing AI Systems" to Serverless Toronto. They've built production AI platforms at scale, and their book focuses on exactly the part that usually gets overlooked: everything required to make an AI system reliable, manageable, and ready for production.
What you'll learn:
- How to ground AI in your company's own data and documents
- Guardrails and safety controls for tool use in production
- Monitoring, cost management, and operational practices that scale
- Quality evaluation: how to measure whether your AI system is actually working
- Production-ready communication patterns and request routing
Why this matters if you're building or buying AI solutions: Every vendor will show you a demo. Few will show you what breaks when real users, real data, and real costs hit the system. Understanding the production layer is how you evaluate vendor claims, scope internal projects realistically, and avoid the "it worked in the notebook" trap.
If you missed Part 1: Val Andrei Fajardo (ex-LlamaIndex founding engineer) built an AI agent from scratch, live, covering tool calling, LLM interfaces, processing loops, and MCP integration. Full recording with timestamped chapters: youtu.be/Q1Co8dplTOI
About the book "Designing AI Systems" is available through Manning's Early Access Program. The book walks you through building a working AI platform with complete Python code in every chapter: provider management, memory, RAG for company data, tool safety, request routing, monitoring, and evaluation. 8 e-book coupons will be raffled at the event.
About the speakers Suhas Suresha is a Senior Machine Learning Engineer at Adobe, where he builds large-scale generative AI platforms across the full machine learning lifecycle. He previously co-founded QALY, where he helped deploy real-time ECG analysis models to more than 100,000 users. He holds a master's degree in computational and applied mathematics from Stanford University.
Dewang Sultania is a Senior Machine Learning Engineer at Netflix, where he designs scalable systems for multimodal generative AI, diffusion models, and video processing. Previously at Adobe, he built production systems for large language models, including data pipelines, fine-tuning, retrieval-based systems, and prompt engineering. He also helped design the platform infrastructure used to deploy LLMs across Adobe's product suite.
Serverless Toronto has been bridging the gap between IT and business needs since 2018. Past sessions: youtube.serverlesstoronto.org
