Software Development Superstream: From AI Conductor to AI Orchestrator
Published by O'Reilly Media, Inc.
Go beyond the prompt to manage autonomous agentic systems
The era of vibe coding is maturing into a disciplined engineering practice. Join Sam Newman and expert guests to explore the critical transition from being a conductor of a single agent to becoming an orchestrator of many. Dive into the emerging agentic software engineering stack to understand how to build systems that plan, test, and execute with high autonomy.
Learn how to spend less time writing code and more time crafting precise technical specifications that serve as the source of truth for autonomous agents. Understand how techniques like spec-driven development, evals, and context engineering can advance your skillset as you move forward in this new agentic world.
What you’ll learn and how you can apply it
- Understand the new trends in agentic engineering and what they mean for your work today
- Work with agents and retain your craft as an engineer
- Build resilient agentic systems that safely ship real work rather than just impressive prototypes
Recommended follow-up:
- Read Building AI Agent Platforms (early release book)
- Read Learning AI-Native Software Engineering (early release book)
- Take Context Engineering for AI-Assisted Coding (live online course with Chelsea Troy)
- Take AI Agents Bootcamp (live online course with Nicole Koenigstein)
Schedule
The time frames are only estimates and may vary according to how the class is progressing.
Introduction – Sam Newman (5 minutes)
Sam Newman welcomes you to the Software Development Superstream.
What Is Loop Engineering? Addy Osmani and Sam Newman Fireside Chat (35 minutes)
Join Sam Newman and Addy Osmani for this fireside chat where they’ll discuss what loop engineering is, why it matters (or maybe why it doesn’t), what this means for agentic development, and what it means for you. Sam and Addy will also answer your questions throughout and in a Q&A at the end of this discussion.
Engineering Craft When Agents Write the Code – Angie Jones (25 minutes)
AI coding agents are changing what it feels like to be a software engineer. That’s challenging engineers’ sense of craftsmanship. If agents are writing the code, is the joy of building lost? Angie Jones, VP of the Agentic AI Foundation, doesn’t think so. In this talk, Angie shares how she found that familiar engineering flow state again, not by writing every line herself, but by designing agentic systems that still require the principles engineers value: systems thinking, decomposition, separation of concerns, state management, etc. Get a prescription for applying the engineering muscles you already have to a new set of building blocks, and get your joy back.
Break (5 minutes)
Redesigning for Agent-Readiness – Akshay Shah and Benjamin Wagner (sponsored by Antithesis) (30 minutes)
Agents are being built to do everything human developers can do, but they’ll go about the work completely differently. What does an agent-ready architecture look like? Benjamin Wagner, CEO at Firebolt.io, recently led a fundamental redesign of Firebolt’s analytical database to make it usable by agents, the developers of the future. In this fireside chat, Akshay Shah, field CTO at Antithesis, asks Benjamin about Firebolt’s experience, what agent-ready design looks like, and how to get there quickly and safely.
This session will be followed by a 30-minute Q&A in a breakout room. Stop by if you have more questions for Akshay and Benjamin.
Using Coding Agents to Build Reliable AI Agents – Alfonso Graziano (35 minutes)
Building production-ready AI agents is a systems problem rather than a simple prompting challenge. Alfonso Graziano, an AI lead at Nearform, introduces a practical, scalable workflow centered around a dual-agent architecture consisting of a coding agent and a product agent: The coding agent is the builder that writes and updates the code base, while the product agent is the custom tool shipped for the end user. Join Alfonso to learn about these agents, the implementation of self-healing evals, and more.
Break (5 minutes)
Beyond Vibe Coding: Building the Harness for Production Agents – Banjo Obayomi (35 minutes)
AI coding agents can produce working software. The harder problem is getting them to produce trustworthy results again and again. In this talk, Banjo Obayomi shows how to build a harness around an agentic workflow using six steps: choose the task, write the rules, capture judgment in files, build checks, review changes with evidence, and earn autonomy gradually. He grounds the method in a daily local events newsletter that gathers, evaluates, verifies, and publishes real information on a schedule. You’ll see how the harness changes as the scope grows, why the bottleneck moves from task clarity to checks, handoffs, and human review, and where engineers should keep control. You’ll leave with a practical, language-agnostic framework for turning one useful agent task into a repeatable production workflow.
Harness Engineering: A Fireside Chat with Sam Newman and Chris Ford (35 minutes)
Most conversations about harness engineering focus on the scaffolding around the model; this one is about the scaffolding around the team. Sam and Thoughtworks’ Chris Ford discuss the organizational harness that determines whether agentic development inside a team thrives or stalls.
Closing Remarks – Sam Newman (5 minutes)
Sam Newman closes out today’s event.
Your Hosts and Selected Speakers
Sam Newman
Sam Newman is a technologist focusing on the areas of cloud, microservices, and continuous delivery—three topics which seem to overlap frequently. He provides consulting, training, and advisory services to startups and large multinational enterprises alike, drawing on his more than 20 years in IT as a developer, sysadmin, and architect. Sam is the author of the best-selling Building Microservices and Monolith to Microservices, both from O’Reilly, and is also an experienced conference speaker.
Addy Osmani
Addy Osmani is an engineering and evangelism leader who spent over 14 years at Google leading developer experience. In recent years his focus shifted to AI: Gemini's developer experience, agentic engineering, coding agents, harnesses, evals and benchmarks, and code quality, work that led into a Director role at Google Cloud AI running Cloud AI Developer Experience and Technical Evangelism, where he helped launch Google's agent developer platform (Agent Platform, Agent CLI, and Agent Studio). Earlier he led Chrome's Developer Experience, working on DevTools, Lighthouse, and Core Web Vitals.
Angie Jones
Angie Jones is vice president of the Agentic AI Foundation where she guides how agentic systems are designed, implemented, and adopted across the global developer ecosystem. An award-winning educator and international keynote speaker, she shares her extensive knowledge with software companies and conference audiences worldwide. An IBM Master inventor, Angie is recognized for her innovative, out-of-the-box thinking, which has led to 27 patented inventions.
Akshay Shah
Akshay Shah is the field CTO at Antithesis, a testing platform to help agents produce bug-free code. In the past, he’s been an infra and platform engineer at Uber, Microsoft, other people’s silly startups, and his own silly startup. Apart from computers, he likes bicycles and coffee.
Benjamin Wagner
Benjamin Wagner is the CEO at Firebolt, where he works with teams on building the fastest and most cost-effective analytical query engine for engineers and agents. Recently, much of that work has focused on synthesizing full database components with coding agents. Benjamin first fell in love with database systems while studying computer science at the Technical University of Munich.
Alfonso Graziano
Alfonso Graziano is an AI lead at Nearform, where he builds production AI agents and helps engineering teams adopt AI-native ways of working. He writes and speaks about context engineering, spec-driven development, and shipping reliable software with AI agents, and has given over 20 talks at conferences across Europe. Alfonso is the author of AI-Native Software Engineering (O’Reilly). He’s based in Italy and is passionate about continuous learning, critical thinking, and problem-solving.
Banjo Obayomi
Banjo Obayomi is a senior solutions architect who helps builders ship with generative AI. He focuses on agentic workflows, developer experience, and the practical craft of building reliable AI-powered systems. Outside of work, he runs a local meetup, builds open source agent tooling, plays Japanese role-playing games, and explores whatever events are happening nearby.
Chris Ford
Chris Ford has worked as a consultant for Thoughtworks for over 15 years, helping clients solve tough problems at the intersection of technology, business, and engineering organization design. He was an early adopter of and contributor to technical developments such as microservices and data mesh. Chris is the author of the upcoming O'Reilly book Agentic Engineering at Scale. His personal technical interests include using functional programming to compose music and pushing the boundaries of the expressiveness of code.
