> Markdown version of [/videos/100003-designing-ux-for-sre-agents-in-high-stakes-incidents?t=263](https://www.wearedevelopers.com/videos/100003-designing-ux-for-sre-agents-in-high-stakes-incidents?t=263). 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).

---

# Designing UX for SRE Agents in High-Stakes Incidents

Why force SREs to navigate conversational mazes during 3 AM system crashes? Discover how the un-shell pattern transforms generative AI into trusted, structural UIs for high-stakes incidents.

- **Speakers:** [Osmar Matos](https://www.wearedevelopers.com/@osmar-matos)
- **Event:** World Congress 2026 Europe
- **Published:** July 9, 2026
- **Duration:** 7:27
- **URL:** https://www.wearedevelopers.com/videos/100003-designing-ux-for-sre-agents-in-high-stakes-incidents

## Summary

The shift from deterministic tools to generative AI fundamentally disrupts how user interfaces are designed, particularly for high-stakes Site Reliability Engineering (SRE). When a system crashes at 3:00 AM, engineers need immediate clarity, not a disjointed conversational maze. While the ultimate goal for AI agents might be a frictionless, UI-free "Jarvis" experience, we are currently navigating a transitional phase that blends deterministic UI elements with generative capabilities. SRE agents must parse logs, metrics, codebases, and ticketing systems to diagnose issues without burying the human operator in overwhelming walls of text.

At Hyground, early chat-based interfaces revealed a steep learning curve; engineers often failed to discover or utilize advanced features without prior training. This led to the adoption of the "un-shell pattern," a design methodology that extracts implicit generative capabilities into familiar, structural UI paradigms. By implementing declarative generative UI rather than purely static or open-ended html rendering, the LLM agent dynamically builds contextual tables, flowcharts, and dashboards. Crucially, the interface exposes the exact server commands the agent orchestrates, maintaining the essential human-in-the-loop steering and trust required in fragile production environments.

This un-shelling approach enables powerful operational mechanics, including unified cross-platform searches across Kubernetes, Jira, and Confluence, as well as dynamic performance dashboards generated directly from intuitive queries. Unlocking these AI SRE capabilities ultimately shifts engineering from purely reactive monitoring to proactive intelligence. Features like automatic Root Cause Analysis (Auto RCA) allow the agent to triage alerts instantly, compiling a complete narrative of what happened, how to solve it now, and how to prevent future regressions. The core design challenge of the future thereby transforms from merely visualizing infrastructure data to actively designing trust and authority with an autonomous intelligence.

**Keywords:** site reliability engineering, generative UI paradigms, autonomous AI agents, human-in-the-loop steering, production incident response, un-shell UI pattern, declarative UI frameworks, automated root cause analysis, auto RCA triage, cross-platform infrastructure search, kubernetes cluster debugging, hyground AI platform, dynamic developer dashboards, LLM agent orchestration, deterministic UI hybrid

## Chapters

1. **Balancing deterministic and generative interfaces in AI applications** (00:03) — Designing predictable user interfaces becomes challenging when generative AI creates dynamic and unpredictable outputs.
1. **Empowering site reliability engineers with integrated AI agents** (01:15) — AI agents embedded directly within production infrastructure help engineers quickly identify the root cause of late-night system failures.
1. **Implementing declarative generative interfaces for incident investigations** (02:21) — Providing specific structural guidelines enables large language models to dynamically generate contextual charts and server command visualizers during troubleshooting.
1. **Extracting AI capabilities into predictable dashboard interfaces** (04:23) — The un-shell pattern translates complex generative chat queries into accessible deterministic dashboards and centralized search functionalities.
1. **Automating root cause analysis through proactive AI investigation** (06:10) — Proactive artificial intelligence systems automatically investigate alerts to present human operators with comprehensive incident summaries and tailored prevention strategies.

## Related Moments

- [Audience Q&A on tooling choices and AI application prototyping](https://www.wearedevelopers.com/videos/100195-xcode-development-redefained) (from "Xcode development redefAIned")
- [Designing user experiences defensively against AI application failures](https://www.wearedevelopers.com/videos/1010-bringing-the-power-of-ai-to-your-application) (from "Bringing the power of AI to your application.")
- [AI agents replacing traditional web interfaces](https://www.wearedevelopers.com/videos/1289-wad-live-22-01-2025-exploring-ai-web-development-and-accessibility-in-tech-with-stefan-judis) (from "WAD Live 22/01/2025: Exploring AI, Web Development, and Accessibility in Tech with Stefan Judis")
- [Integrating human engineers and autonomous agents in the SDLC](https://www.wearedevelopers.com/videos/100106-craftsmanship-in-the-age-of-agents) (from "Craftsmanship in the Age of Agents")
- [The shift from predictable UX to adaptive AI experiences](https://www.wearedevelopers.com/videos/2122-designing-the-intelligence-layer-the-future-of-products-beyond-interfaces) (from "Designing the intelligence layer: The future of products beyond interfaces")
- [Summarizing developer experience and artificial intelligence companions](https://www.wearedevelopers.com/videos/884-forget-developer-platforms-think-developer-productivity) (from "Forget Developer Platforms, Think Developer Productivity!")

## Related Articles

- [WWC24 Talk - Scott Hanselman - AI: Superhero or Supervillain?](https://www.wearedevelopers.com/magazine/469-wwc24-talk-scott-hanselman-ai-superhero-or-supervillain)
- [Never delegate the understanding](https://www.wearedevelopers.com/magazine/749-never-delegate-the-understanding)
- [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)

## Related Jobs

- [Senior Engineer, Infrastructure Platform](https://www.wearedevelopers.com/jobs/ext/328836-senior-engineer-infrastructure-platform) at **Intercom, Inc.**
- [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 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**
- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [AI Operations Manager (all genders)](https://www.wearedevelopers.com/jobs/48263-ai-operations-manager-all-genders) at **envelio**