diff --git a/events/2026-05-20-virtual-webinar/README.md b/events/2026-05-20-virtual-webinar/README.md
new file mode 100644
index 0000000..9a40b33
--- /dev/null
+++ b/events/2026-05-20-virtual-webinar/README.md
@@ -0,0 +1,156 @@
+# Webinar: Autonomous Software Engineering with Devin
+
+## Event Details
+
+| | |
+|---|---|
+| **Date** | 2026-05-20 |
+| **Location** | Virtual |
+| **Host Organization** | *(customer)* |
+| **Duration** | 1 hour |
+| **Format** | Webinar — presentation with live walkthrough |
+| **Audience** | Engineering leadership, architects, delivery leads, and developers exploring AI-augmented development |
+| **Event Site** | TBD |
+
+## Abstract
+
+Asynchronous AI is accelerating the software development lifecycle. This one-hour webinar introduces autonomous software engineering and shows Devin in action. Attendees will explore real-world use cases across the SDLC, understand where autonomous AI drives the most impact, and see how organizations are beginning to integrate these capabilities into development workflows.
+
+The session includes a live walkthrough, practical adoption considerations, and emerging best practices. Participants will leave with a clear view of how asynchronous AI can augment engineering teams and where it fits within their technology strategy.
+
+---
+
+## Agenda
+
+| Time | Activity | Notes |
+|------|----------|-------|
+| 0:00 | Welcome and introductions | Set context: what is autonomous software engineering? |
+| 0:05 | **The Autonomous Engineering Model** | How Devin works: cloud-based agent, event-driven triggers, clean-room execution |
+| 0:15 | **Real-World Use Cases Across the SDLC** | Event-driven (incidents, security findings, tickets), large-scale campaigns (migrations, modernization), ongoing O&M (dependency hygiene, test coverage) |
+| 0:30 | **Live Walkthrough** | Devin session in action — show the prompt → execution → PR → feedback loop |
+| 0:45 | **Adoption Considerations & Best Practices** | Where to start, design patterns for success, team integration model |
+| 0:55 | Q&A and next steps | |
+
+---
+
+## Narrative Arc
+
+### 1. The Autonomous Engineering Model (10 min)
+
+Position Devin as an on-demand, cloud-based software engineering agent:
+
+- **Event-driven triggers** — Devin responds to signals from existing toolchain (outage incidents, security findings, ticket assignments, CI failures)
+- **Large-scale campaigns** — Codebase modernization, language/framework migration and translation to reduce licensing costs, bulk remediation
+- **Clean-room execution** — Devin starts with access to nothing; organizations grant scoped credentials (service accounts, API keys) for exactly the systems Devin needs
+- **Context retrieval** — Devin retrieves context from indexed codebases and programmatically accesses remote resources via its shell and MCP integrations
+
+Reference: [shared/general-themes/when-to-use-devin.md](../../shared/general-themes/when-to-use-devin.md), [shared/general-themes/architecture-strengths.md](../../shared/general-themes/architecture-strengths.md)
+
+### 2. Real-World Use Cases (15 min)
+
+Walk through concrete scenarios where autonomous AI drives impact:
+
+| Category | Examples | Value |
+|----------|----------|-------|
+| **Incident response** | Alert fires → Devin queries logs via MCP → identifies root cause → opens fix PR | Reduced MTTR without engineer context-switch |
+| **Security remediation** | SAST scan detects CVEs → Devin remediates → re-scan verifies → closed loop | Vulnerability exposure window: sprint cycles → hours |
+| **Migration at scale** | COBOL → Java, SAS → Python/Snowflake, Spring Boot 2 → 3 across N services | Child agents parallelize; weeks become days |
+| **Ongoing O&M** | Scheduled dependency bumps, dead code cleanup, documentation refresh | Engineers focus on architecture; Devin handles hygiene |
+| **Test generation** | Coverage gaps identified → Devin generates tests following existing conventions | Quality improvement without sprint capacity cost |
+
+Emphasize the sweet spot: well-defined work that wouldn't otherwise get done due to constrained engineering capacity, prohibitive volume, insufficient priority, or engineers having higher-value tasks to focus on.
+
+Reference: [shared/general-themes/when-to-use-devin.md](../../shared/general-themes/when-to-use-devin.md)
+
+### 3. Live Walkthrough (15 min)
+
+Show Devin in action end-to-end. Suggested options (pick one based on audience):
+
+**Option A — Gap Analysis (low-risk, analysis-only):**
+- Paste a prompt asking Devin to analyze a codebase and produce a technical assessment
+- Show Devin planning, reading code, producing structured documentation
+- Show the resulting PR with the analysis artifacts
+
+**Option B — Security Remediation (event-driven pattern):**
+- Show a vulnerability scan result triggering a Devin session
+- Show Devin reading the findings, implementing fixes, running verification
+- Show the closed-loop: scan → fix → re-scan → PR
+
+**Option C — Migration (large-scale pattern):**
+- Show a parent agent identifying migration targets
+- Show child agents spawning to handle individual targets in parallel
+- Show multiple PRs being created simultaneously
+
+During the walkthrough, highlight:
+- The PR feedback loop (comments → Devin iterates)
+- CI monitoring (Devin watches checks, fixes failures)
+- Hibernation/resume (Devin sleeps when waiting, wakes on new input)
+
+### 4. Adoption Considerations & Best Practices (10 min)
+
+**Design patterns for success:**
+- Make code locally buildable and testable — Devin verifies its own work by running your build/test suite
+- Connect Devin to event sources (CI, alerting, issue trackers) for autonomous operation
+- Invest in the context layer (knowledge notes, environment config, MCP connections) — compounds over time
+- Start with well-defined, low-risk tasks; expand scope as confidence grows
+
+**Platform capabilities that compound:**
+- **Scheduled sessions** — Proactive O&M (dependency bumps, compliance audits) without human initiation
+- **Playbooks** — Reusable methodologies that encode proven approaches; any team member can invoke them
+- **Child agents** — Divide and conquer at scale with a proven, repeatable methodology
+- **Team-based configuration** — Shared context layer and automations; configure once, benefit everywhere
+- **Devin Review** — Proactively find bugs in new PRs, help humans digest large PRs, and proactively remediate discovered issues
+
+**Collaboration model:**
+- Devin is always waiting for feedback from PR comments or CI checks
+- Multiple team members can communicate with the same Devin agent
+- Devin hibernates its VM after inactivity but resumes from hibernated state upon new feedback to continue work
+
+Reference: [shared/general-themes/design-patterns-for-devin.md](../../shared/general-themes/design-patterns-for-devin.md), [shared/general-themes/platform-capabilities.md](../../shared/general-themes/platform-capabilities.md), [shared/general-themes/collaboration-model.md](../../shared/general-themes/collaboration-model.md)
+
+---
+
+## Key Takeaways
+
+- **"Unlock work that is currently blocked or deferred"** — Devin handles well-defined tasks that wouldn't get done due to constrained capacity, difficulty, or volume
+- **"Event-driven and always-on"** — Devin responds to incidents, findings, and tickets in real time without engineer context-switching
+- **"Verify its own work"** — locally initiated testing, CI monitoring, and PR-based review create a closed feedback loop
+- **"Compounds over time"** — scheduled sessions, playbooks, knowledge, and team configuration make Devin more effective with each iteration
+- **"Augments, not replaces"** — engineers focus on architecture, design, and judgment; Devin handles implementation at scale
+
+---
+
+## Facilitator Notes
+
+### Preparation
+- Pre-stage 1-2 Devin sessions that can be shown live (gap analysis or security remediation work well for live walkthroughs)
+- Have the general-themes documents available as reference for narrative consistency
+- Prepare 2-3 customer-relevant use cases based on the audience's tech stack and pain points
+
+### Audience Adaptation
+
+| Audience | Lead With | Emphasize |
+|----------|-----------|-----------|
+| Engineering leadership | Capacity unlocking, velocity multiplication | ROI framing, team augmentation (not replacement) |
+| Security / compliance | Event-driven SAST remediation, closed-loop verification | Risk reduction, vulnerability exposure window |
+| Architects | Clean-room execution, context retrieval, design patterns | System integration, MCP connectivity |
+| Delivery / program management | Large-scale campaigns, child agents | Timeline compression, parallelization |
+
+### Follow-Up Options
+
+After the webinar, offer participants next steps:
+
+1. **Hands-on workshop** — 2-4 hour session where participants use Devin directly (see other events in this catalog)
+2. **Proof of concept** — Scoped engagement against the customer's own codebase
+3. **Platform trial** — Team access to Devin with onboarding support
+
+---
+
+## Related Resources
+
+| Resource | Link |
+|----------|------|
+| General themes | [shared/general-themes/](../../shared/general-themes/) |
+| Value narratives | [shared/general-themes/value-narratives.md](../../shared/general-themes/value-narratives.md) |
+| Hands-on workshop examples | [events/](../) |
+| Facilitator guide | [shared/facilitator-guide.md](../../shared/facilitator-guide.md) |