> Markdown version of [/videos/100238-how-building-with-ai-can-double-the-throughput-of-your-engineering-team?t=664](https://www.wearedevelopers.com/videos/100238-how-building-with-ai-can-double-the-throughput-of-your-engineering-team?t=664). 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).

---

# How building with AI can double the throughput of your engineering team

Facing plunging growth, Intercom didn't just launch a chatbot. They treated AI like a senior engineer with deep system access—and tripled their R&D throughput in 16 months.

- **Speakers:** [Brian Scanlan](https://www.wearedevelopers.com/@brian-scanlan)
- **Event:** World Congress 2026 Europe
- **Published:** July 10, 2026
- **Duration:** 30:50
- **URL:** https://www.wearedevelopers.com/videos/100238-how-building-with-ai-can-double-the-throughput-of-your-engineering-team

## Summary

Intercom, a 15-year-old B2B SaaS startup, successfully hard-pivoted into an AI-first company following the release of ChatGPT. Facing collapsing growth rates, leadership made a bold decision not only to launch an AI customer support agent (Finn) but to fundamentally transform internal operations. To achieve this, leadership mandated a sweeping goal to double the engineering team's R&D throughput without compromising security, performance, or quality.

Instead of allowing fragmented tool adoption or focusing on performative metrics like token maxing, Intercom implemented a unified platform strategy. They went all-in on Claude Code, treating the AI like a new senior engineer. The platform team provided the AI with comprehensive access to internal systems—including Datadog, Honeycomb, and internal admin tools—and onboarded it with strict coding conventions and linting rules. Crucially, a full-time, highly skilled team of staff and principal engineers was dedicated to supporting developers, creating small, composable skills that act as building blocks for compounding productivity.

Rather than giving AI micro-tasks, Brian emphasizes the value of assigning AI high-level problems, enabling it to determine which tools to use and independently navigate complex incidents. This agent-first approach successfully tripled engineering throughput in 16 months, decimated long-standing defect backlogs, and significantly reduced the total cost of ownership per pull request. To continuously improve, Intercom stores anonymized AI session telemetry for deep analysis. Ultimately, this shift is completely reshaping traditional work structures, unlocking experimental manager-less teams where engineers seamlessly adapt into product and design roles.

**Keywords:** engineering throughput optimization, developer productivity metrics, ai tooling adoption, agent-first sdlc, claude code implementation, pull request automation, ai platform strategy, defect backlog reduction, manager-less engineering teams, ai telemetry analysis, software development transformation, ai-assisted code review, total cost of ownership per pr, agent-first architecture, engineering enablement team

## Chapters

1. **Intercom's strategic pivot to an AI-first software company** (01:48) — How Intercom shifted its entire business model to focus on AI products following the release of ChatGPT.
1. **Setting the ambitious goal to double engineering throughput** (07:03) — Mandating a two-fold increase in production deployments and real feature changes without compromising quality or security.
1. **Achieving and sustaining three times engineering productivity** (09:35) — Reaching the productivity goal months early and maintaining the momentum across the entire research and development workflow.
1. **Driving organizational change through clear engineering mandates** (11:04) — How leadership enforced AI adoption by updating engineering expectations, offering rewards, and providing dedicated enablement teams.
1. **Committing to a unified foundational platform strategy** (14:03) — Choosing a single AI platform and integrating it deeply rather than fragmenting efforts across multiple tools and models.
1. **Integrating AI agents as senior engineering team members** (16:37) — Granting Claude full access to internal tools and onboarding it with coding standards, architecture rules, and testing guidelines.
1. **Focusing developer effort on durable evergreen skills** (19:26) — Prioritizing the unblocking of knowledge, context, and environment access over building transient tool-specific workflows.
1. **Prompting agents with high-level problems instead of tasks** (21:07) — Providing AI agents with broad problem contexts empowers them to autonomously determine the necessary skills and actions for resolution.
1. **Measuring developer experience and the cost of output** (23:16) — Tracking positive trends in developer satisfaction alongside a plunging total cost of ownership per code change.
1. **Eliminating legacy defect backlogs and accelerating feature delivery** (24:54) — Using AI to reach zero defects and significantly halve the time required to ship real features to customers.
1. **Composing small tasks into powerful open-source skills** (26:54) — Building composite automation skills for specific development tasks to create comprehensive workflows like thermonuclear code reviews.
1. **Capturing AI session telemetry for deep usage insights** (27:58) — Recording and analyzing agent session transcripts to identify gaps and improve AI integration across the engineering lifecycle.
1. **Expanding AI agents across workflows and team structures** (28:59) — Deploying dedicated agents for tasks like deployment monitoring while experimenting with manager-less, cross-functional engineering roles.

## Related Moments

- [Shifting developer workloads and realistic AI productivity gains](https://www.wearedevelopers.com/videos/1830-wearedevelopers-live-speculaitions) (from "WeAreDevelopers LIVE - SpeculAItions")
- [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!")
- [Inspiration and challenges of scaling AI agent communication](https://www.wearedevelopers.com/videos/1465-event-driven-architecture-breaking-conversational-barriers-with-distributed-ai-agents) (from "Event-Driven Architecture: Breaking Conversational Barriers with Distributed AI Agents")
- [Using artificial intelligence to reimagine developer experience](https://www.wearedevelopers.com/videos/1546-ai-pair-programming-with-github-copilot-at-sap-looking-back-looking-forward) (from "AI Pair Programming with GitHub Copilot at SAP: Looking Back, Looking Forward!")
- [Leveraging AI efficiencies to scale development output](https://www.wearedevelopers.com/videos/1405-fireside-chat-with-werner-vogels-vp-cto-amazon-com-daniel-gebler-cto-at-picnic) (from "Fireside Chat with Werner Vogels, VP & CTO, Amazon.com & Daniel Gebler, CTO at Picnic")
- [Evaluating direct and long-term AI return on investment](https://www.wearedevelopers.com/videos/100253-ai-in-high-stakes-industries-lessons-learned) (from "AI in High-Stakes Industries: Lessons Learned")

## Related Articles

- [Never delegate the understanding](https://www.wearedevelopers.com/magazine/749-never-delegate-the-understanding)
- [Stephan Gillich - Bringing AI Everywhere](https://www.wearedevelopers.com/magazine/489-stephan-gillich-bringing-ai-everywhere)
- [Navigating the AI Shift](https://www.wearedevelopers.com/magazine/629-navigating-the-ai-shift)
- [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care)

## Related Jobs

- [Senior Engineer, Infrastructure Platform](https://www.wearedevelopers.com/jobs/ext/328836-senior-engineer-infrastructure-platform) at **Intercom, Inc.**
- [Senior Forward Deployed Engineer](https://www.wearedevelopers.com/jobs/ext/1915768-senior-forward-deployed-engineer) at **Intercom**
- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [Principal Product Manager, Agent Platform](https://www.wearedevelopers.com/jobs/ext/277541-principal-product-manager-agent-platform) at **GitHub**
- [Principal Field Architect - AI Agents](https://www.wearedevelopers.com/jobs/ext/1442858-principal-field-architect-ai-agents) at **Twilio**
- [Senior Web Designer, Growth](https://www.wearedevelopers.com/jobs/ext/102722-senior-web-designer-growth) at **Intercom, Inc.**