> Markdown version of [/videos/100166-shipping-with-confidence-observability-and-quality-at-scale?t=188](https://www.wearedevelopers.com/videos/100166-shipping-with-confidence-observability-and-quality-at-scale?t=188). 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).

---

# Shipping with Confidence: Observability and Quality at Scale

AI generates code faster than humanity can safely review it. Discover how to evolve your observability practices to balance agentic release velocity with uncompromised structural integrity.

- **Speakers:** [Christian Heilmann](https://www.wearedevelopers.com/@christian-heilmann), [Mirko Novakovic](https://www.wearedevelopers.com/@mirko-novakovic), [Prince Kohli](https://www.wearedevelopers.com/@prince-kohli), [Daniel Gebler](https://www.wearedevelopers.com/@daniel-gebler)
- **Event:** World Congress 2026 Europe
- **Published:** July 9, 2026
- **Duration:** 31:24
- **URL:** https://www.wearedevelopers.com/videos/100166-shipping-with-confidence-observability-and-quality-at-scale

## Summary

As systems grow more complex and generative AI exponentially accelerates code creation, the historically manageable gap between writing and verifying software has fractured. Engineering leaders are now confronting a landscape where developers generate code faster than humanity can safely review it, contributing to a massive surge in production incidents per pull request. The panel underscores how the disappearance of dedicated QA departments forces a profound operational pivot; while AI empowers developers to ostensibly ‘own the full stack,’ it simultaneously burdens them with unstructured, overwhelming output governance.

A core insight from the discussion is that quality assurance is morphing from checking outputs to perfecting inputs. Because AI is increasingly capable of translating precise specifications into execution, the true bottleneck has become humanity's inability to describe intent without ambiguity. Furthermore, while code generation 'shifts left,' the sheer volume pushes a 'shift right' effect, creating severe human review bottlenecks and forcing teams to catch errors in production via risk scoring. To cope, observability practices must evolve beyond simple request-response metrics. Teams must effectively monitor long-running agentic workflows—tracking thousands of LLM calls, evaluating reasoning accuracy, and running digital twin 'dry runs' before allowing an AI agent to execute physical commands in hardware environments like automated logistics.

Maintaining confidence at scale ultimately requires strict boundaries around human authority and security. The consensus dictates that an engineer must act as a structured supervisor orchestrating AI workers, a reality emphasized by the panel: "These are my agents... if I'm going to say, 'Yes, ship it,' then I better be sure I understand what they do." Organizations are heavily warned against assigning code ownership to personnel who can prompt generative code but lack the architectural knowledge to fix underlying issues. By adopting AI-based intent verification, utilizing separated models to prevent testing 'echo chambers,' and enforcing blast radius impact scoring, teams can successfully balance agentic release velocity with structural integrity.

**Keywords:** ai code generation velocity, production incident monitoring, shift-left quality assurance, intent-based software verification, agentic workflow observability, digital twin dry runs, llm token optimization, model context protocol implementations, hardware-software continuous delivery, human-in-the-loop ai governance, code ownership boundaries, automated software rollback mechanisms, supply chain security patching, generative ai blast radius scoring, long-running agent evaluations

## Chapters

1. **Balancing rapid rollbacks in physical and digital supply chains** (01:58) — Operating automated warehouses requires specialized simulation and testing to sync digital states with the physical world during system rollbacks.
1. **Bridging the gap between AI code generation and verification** (03:08) — The exponential speed of AI code creation outpaces traditional testing methods and demands intent-based testing to maintain overall productivity.
1. **Shifting responsibility and the loss of dedicated quality assurance** (04:59) — As conventional testing departments disappear, engineers face increasing pressure to own the entire tech stack and independently maintain trust in their output.
1. **Handling the sudden rise of production incidents and risks** (06:27) — With the dramatic increase in production bugs due to faster coding cycles, organizations must implement scoring and blast radius measurements to qualify pull requests.
1. **Improving requirement specifications to ensure reliable AI code translations** (08:37) — Translating software requirements into functioning syntax requires professionals to describe inputs more explicitly rather than relying solely on output verification.
1. **Managing agent token usage and tracking return on investment** (12:45) — The widespread experimentation with modern coding tools necessitates separating models for generation versus testing to optimize prompt costs effectively.
1. **Balancing rapid software updates against supply chain attack risks** (15:02) — Development organizations struggle to find scalable practices for safely upgrading external dependencies without falling victim to cascading zero-day module compromises.
1. **Pacing release cycles for digital platforms versus hardware infrastructure** (17:13) — Fast daily deployments work well for consumer shopping applications but must be deliberately slowed down for complex robotics and physical logistics updates.
1. **Adapting observability strategies for long-running enterprise AI agents** (18:55) — Monitoring autonomous entities requires shifting from fast request-response metrics to tracking extensive agentic workflows and mathematically evaluating output accuracy.
1. **Structuring human governance over autonomous enterprise software workflows** (20:43) — Large companies are maturing their AI adoption by positioning software engineers as active supervisors over fully orchestrated, intelligent pipelines.
1. **Establishing secure environments for dry-running complex autonomous agent actions** (22:27) — Validating complex physical operations involves letting algorithms simulate scenarios before returning quantitative performance measurements for operators to approve in the real world.
1. **Defining developer ownership within highly distributed agentic software architectures** (24:11) — Engineers must understand underlying system pathways deeply enough to claim total accountability for any syntax generated by their autonomous tooling.
1. **Closing the development loop with automated AI incident responses** (26:49) — Managing the escalating velocity of coding pipelines requires deploying analytical agents in production to quickly diagnose anomalies and submit automated repairs.
1. **Selecting appropriately when to shift testing left versus right** (28:05) — Rather than reflexively front-loading every system check, deployment teams should distribute verification processes appropriately based on the precise cost of late correction.

## Related Moments

- [Navigating developer bottlenecks and human accountability](https://www.wearedevelopers.com/videos/100265-fireside-chat-in-conversation-with-werner-vogels-cto-of-amazon-com) (from "Fireside Chat - In conversation with Werner Vogels, CTO of Amazon.com")
- [Security integration and AI skepticism in developer tooling](https://www.wearedevelopers.com/videos/1830-wearedevelopers-live-speculaitions) (from "WeAreDevelopers LIVE - SpeculAItions")
- [Managing the operational impact of AI-generated code volume](https://www.wearedevelopers.com/videos/100332-software-that-fixes-itself) (from "Software That Fixes Itself")
- [Shifting developer workloads and realistic AI productivity gains](https://www.wearedevelopers.com/videos/1830-wearedevelopers-live-speculaitions) (from "WeAreDevelopers LIVE - SpeculAItions")
- [Managing AI speed and the rise of verification debt](https://www.wearedevelopers.com/videos/100265-fireside-chat-in-conversation-with-werner-vogels-cto-of-amazon-com) (from "Fireside Chat - In conversation with Werner Vogels, CTO of Amazon.com")
- [Balancing platform availability and security with AI feature releases](https://www.wearedevelopers.com/videos/100106-craftsmanship-in-the-age-of-agents) (from "Craftsmanship in the Age of Agents")

## Related Articles

- [Exploring AI: Opportunities and Risks for Developers](https://www.wearedevelopers.com/magazine/522-exploring-ai-opportunities-and-risks-for-developers)
- [What is Software Engineering in the Age of AI?](https://www.wearedevelopers.com/magazine/640-what-is-software-engineering-in-the-age-of-ai)
- [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care)
- [Transforming Software Development: The Role of AI and Developer Tools](https://www.wearedevelopers.com/magazine/527-transforming-software-development-the-role-of-ai-and-developer-tools)

## Related Jobs

- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [Senior AI Agent Software Engineer (Go, Python) (m/f/x)](https://www.wearedevelopers.com/jobs/48277-senior-ai-agent-software-engineer-go-python-m-f-x) at **Dynatrace**
- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Principal Product Manager, Agent Platform](https://www.wearedevelopers.com/jobs/ext/277541-principal-product-manager-agent-platform) at **GitHub**
- [Senior Engineer, Infrastructure Platform](https://www.wearedevelopers.com/jobs/ext/328836-senior-engineer-infrastructure-platform) at **Intercom, Inc.**