> Markdown version of [/videos/100376-the-future-is-calling-but-you-re-busy-prompting?t=217](https://www.wearedevelopers.com/videos/100376-the-future-is-calling-but-you-re-busy-prompting?t=217). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# The Future Is Calling But You're Busy Prompting

Stop fighting to be the human in the loop and start engineering the loop. Discover how to architect autonomous multi-agent systems using foundational software engineering principles.

- **Speakers:** [Angie Jones](https://www.wearedevelopers.com/@angie-jones)
- **Event:** World Congress 2026 North America
- **Published:** September 26, 2026
- **Duration:** 31:55
- **URL:** https://www.wearedevelopers.com/videos/100376-the-future-is-calling-but-you-re-busy-prompting

## Summary

Developers are currently hyper-focused on learning to prompt and consume AI coding agents, but the real future of software engineering lies in building agentic systems. Using a real-world scenario of a 2 AM flight cancellation handled entirely by autonomous agents, the presentation illustrates the complex, multi-agent orchestration required to make this a reality. The transition requires moving from fighting to be the "human in the loop" to actually "engineering the loop."<br><br>Building these non-deterministic systems relies heavily on foundational software engineering principles, such as systems thinking, state management, and separation of concerns. Developers must protect agent context windows through progressive disclosure, decide when to use standard code over LLMs via zero token architecture, and implement robust threat modeling against prompt injections. Essential new skills include context engineering using standards like the Model Context Protocol (MCP) and designing agentic memory to retain user preferences without polluting future tasks.<br><br>As agents evolve to operate on behalf of users—navigating budgets, booking APIs, and making autonomous decisions—developers must master new paradigms like agent-to-agent (A2A) communication and delegation via "agent cards." Because these systems are stateless and unpredictable, designing for API idempotency and implementing deep AI observability are non-negotiable. Ultimately, software engineering is far from dead; "using agents is only the beginning, our future is in building them."

**Keywords:** agentic software engineering, autonomous agent orchestration, model context protocol (MCP), zero token architecture, agent-to-agent communication (A2A), agentic memory design, LLM state management, prompt injection mitigation, API idempotency, AI context engineering, separation of concerns, non-deterministic system testing, agentic loop management, AI observability

## Chapters

1. **Real-world scenarios for autonomous agentic systems** (01:06) — An automated travel agent handles flight cancellations by coordinating rebooking and payments.
1. **Unpacking the technical challenges of agentic commerce** (03:37) — Purchasing tickets autonomously requires standards for merchants to verify buyer intent and process payments securely.
1. **Shifting focus from consuming to building agentic software** (05:44) — Engineers must move beyond simply using coding assistants to designing and developing agentic systems themselves.
1. **Designing the core loop of non-deterministic agents** (08:09) — Agentic loops require providing models with the necessary information and tools to make and act on decisions iteratively.
1. **Applying systems thinking to clarify agent responsibilities** (09:49) — Defining clear boundaries ensures that agents only perform authorized tasks within the broader application architecture.
1. **Protecting context windows through progressive information disclosure** (11:00) — Modularizing tasks into specific skills and sub-agents prevents system prompts from overloading the context window.
1. **Selecting the appropriate models for specific architectural components** (13:02) — Evaluating tasks determines whether to use generative models, decision models, or standard code for optimal efficiency.
1. **Implementing state management and idempotency in stateless systems** (14:41) — Robust state management prevents chaotic duplicate actions when tool calls fail or systems restart.
1. **Threat modeling to prevent malicious prompt injection attacks** (15:59) — Restricting tool access and explicitly handling external inputs protects agents from unauthorized manipulations.
1. **Enforcing time and budget constraints within agentic loops** (17:31) — Building hard limits into the surrounding software ensures that autonomous processes remain efficient and controlled.
1. **Engineering context through standard connection protocols** (19:11) — Integrating internal databases and external systems via standard protocols gives agents the necessary context to make accurate decisions.
1. **Designing specialized skills and managing agentic memory** (21:09) — Defining clear processes for specific tasks and determining what information to retain shapes how agents handle future scenarios.
1. **Designing tool interfaces and managing execution permissions** (24:25) — Providing clear tool descriptions and establishing approval workflows allows models to select actions within authorized boundaries.
1. **Evaluating non-deterministic agentic systems through simulated scenarios** (26:07) — Testing autonomous software requires running repeated simulations to measure how consistently the system meets defined constraints.
1. **Coordinating tasks across multiple third-party autonomous agents** (27:49) — Leveraging agent cards and communication standards enables seamless task delegation between systems built by different organizations.
1. **Implementing observability for tracing non-deterministic execution flows** (29:56) — Capturing available information at runtime enables developers to trace back errors and fix gaps in autonomous processes.
1. **The future of software engineering in autonomous systems** (30:51) — The transition toward agentic software creates new opportunities for engineers to build reliable and efficient autonomous applications.

## Related Moments

- [The evolving role of software engineers alongside agents](https://www.wearedevelopers.com/videos/100132-the-agent-interface-layer-protocols-tools-and-trust-boundaries) (from "The Agent Interface Layer: Protocols, Tools and Trust Boundaries")
- [Preserving engineering fundamentals in agentic development](https://www.wearedevelopers.com/videos/1897-agents-version-control-and-bunnies-daniel-siegl-david-payr) (from "Agents, Version Control and Bunnies - Daniel Siegl & David Payr")
- [Introduction to agentic engineering and talk overview](https://www.wearedevelopers.com/videos/100517-the-agentic-engineering-loop) (from "The Agentic Engineering Loop")
- [The evolution of AI programming and agentic workflows](https://www.wearedevelopers.com/videos/100032-under-the-hood-of-building-on-lovable) (from "Under the Hood of Building on Lovable")
- [Balancing developer autonomy with the adoption of coding agents](https://www.wearedevelopers.com/videos/100198-the-last-mile-of-ai-from-prototype-to-production) (from "The Last Mile of AI: From Prototype to Production")
- [Adapting product design for proactive and agentic AI](https://www.wearedevelopers.com/videos/2062-designing-ai-first-products-when-the-interface-is-no-longer-the-product) (from "Designing AI-first products: When the interface is no longer the product")

## Related Articles

- [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care)
- [Exploring AI: Opportunities and Risks for Developers](https://www.wearedevelopers.com/magazine/522-exploring-ai-opportunities-and-risks-for-developers)
- [WWC24 Talk - Scott Hanselman - AI: Superhero or Supervillain?](https://www.wearedevelopers.com/magazine/469-wwc24-talk-scott-hanselman-ai-superhero-or-supervillain)
- [What is Software Engineering in the Age of AI?](https://www.wearedevelopers.com/magazine/640-what-is-software-engineering-in-the-age-of-ai)

## Related Jobs

- [Senior AI Developer](https://www.wearedevelopers.com/jobs/ext/2836034-senior-ai-developer) at **PwC**
- [Staff Software Engineer, Agentic Platform](https://www.wearedevelopers.com/jobs/48464-staff-software-engineer-agentic-platform) at **Docker, Inc.**
- [Principal AI Forward Deployed Engineer](https://www.wearedevelopers.com/jobs/48531-principal-ai-forward-deployed-engineer) at **Expedia**
- [Principal Software Engineer (m/f/x)](https://www.wearedevelopers.com/jobs/48548-principal-software-engineer-m-f-x) at **Dynatrace**
- [Head of Agentic AI / Lead AI Engineer](https://www.wearedevelopers.com/jobs/48488-head-of-agentic-ai-lead-ai-engineer) at **1st solution consulting gmbh**
- [Senior AI/ML Engineer](https://www.wearedevelopers.com/jobs/48352-senior-ai-ml-engineer) at **PagerDuty**