> Markdown version of [/videos/1893-agents-and-ai-in-enterprise-dona-sarkar-patrick-chanezon?t=1252](https://www.wearedevelopers.com/videos/1893-agents-and-ai-in-enterprise-dona-sarkar-patrick-chanezon?t=1252). 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).

---

# Agents and AI in Enterprise - Dona Sarkar & Patrick Chanezon

Dona Sarkar and Patrick Chanezon reveal that engineers are officially evolving into agent bosses. Discover why the CLI is back and how enterprise AI kills billable hours.

- **Speakers:** Dona Sarkar, Patrick Chanezon
- **Event:** Coffee With Developers
- **Published:** May 13, 2026
- **Duration:** 35:23
- **URL:** https://www.wearedevelopers.com/videos/1893-agents-and-ai-in-enterprise-dona-sarkar-patrick-chanezon

## Summary

The integration of AI agents in enterprise environments is forcing a fundamental shift in how developers and developer relations (DevRel) teams operate. Engineers are transitioning from individual contributors to "agent bosses" who manage automated coding assistants rather than writing every line of code themselves. This shift positions agents as a new kind of consumer for developer tools, requiring specific APIs and robust sandboxing frameworks. Surprisingly, the command-line interface (CLI) is re-emerging as the primary form factor for managing these systems, providing a structured, controllable environment for task automation—though balancing strict permissions with unrestricted agent capabilities remains a significant security challenge.

As tools successfully function like an "untrained junior engineer," organizations must completely rethink tech talent mentorship, shifting focus from raw code generation to architectural oversight and rigorous code review. Furthermore, relying too heavily on generative tools can lead to average, predictable results—described as "mid" output by the speakers—putting professionals at risk of losing their unique human voice. Simultaneously, the enterprise AI landscape is exiting its artificially inflated honeymoon phase. Moving away from the wasteful developer trend of "token maxing," companies are rigorously evaluating the financial and environmental toll of running advanced models. Tech leaders are now tracking specialized efficiency metrics like "token per dollar per watt" and utilizing the Software Carbon Intensity (SCI) framework. This cost-conscious reality is driving a return to localized client-server architectures, where open-source models handle straightforward local tasks while selectively routing complex queries to larger cloud models.

For established enterprises constrained by legacy systems, meaningful AI implementation requires targeting entirely new, "impossible problems"—such as zero-waste supply chains in fast fashion—rather than aggressively retrofitting AI into decades-old technology stacks. Because AI radically accelerates task completion, service industries are being forced to shift from activity-based measurements (like billable hours or in-app engagement metrics) to outcome-based pricing models. Ultimately, successful enterprise AI adoption demands concrete corporate guardrails, extensive user testing, and a mandate to solve genuine operational pain points rather than pursuing shiny features for investor optics.

**Keywords:** enterprise ai adoption strategies, developer relations ai automation, ai agent container sandboxing, command line ai interfaces, llm token cost management, software carbon intensity measurement, hybrid local llm architecture, junior developer ai mentorship, ai assisted code reviews, outcome based ai pricing, enterprise legacy modernization, measuring ai productivity, automated devrel workflows, ai agent security permissions

## Chapters

1. **Microsoft developer relations and the origins of AgentCon** (00:00) — Hosting specialized conferences helps scale technical knowledge about autonomous agents to global developer communities.
1. **Transitioning developers from individual contributors to agent bosses** (02:47) — Treating artificial intelligence systems as team members fundamentally changes the traditional engineering workflow.
1. **Automating developer workflows using command line interface agents** (04:20) — Integrating command line interfaces allows developer relations teams to automate administrative community programs.
1. **Preserving authentic human voice in developer conference submissions** (05:41) — Writing conference proposals without automation preserves speaker personalities and builds genuine professional relationships.
1. **Managing security and sandboxing for command line agents** (09:04) — Granting models complete system access creates security challenges that require proper sandbox environments.
1. **Navigating model guardrails and token usage in enterprise** (12:31) — Providing structured corporate environments allows engineers to leverage foundational models without manually managing tokens.
1. **Shifting focus from token maxing to AI cost management** (16:41) — Measuring computational efficiency and software carbon intensity optimizes the return on hardware investments.
1. **Returning to client-server architectures with local open source models** (18:51) — Running smaller open-source models directly on endpoints improves application privacy and reduces networking dependencies.
1. **Redesigning junior developer training in the artificial intelligence era** (20:52) — Pairing novices with senior staff to review automated pull requests accelerates engineering onboarding.
1. **Solving impossible enterprise problems instead of retrofitting applications** (25:25) — Focusing on previously unsolvable industry challenges prevents organizations from unnecessarily breaking functional legacy systems.
1. **Measuring artificial intelligence success through improved business outcomes** (29:56) — Transitioning corporate metrics from raw execution counts to actual productivity value accurately tracks technology investments.
1. **Validating product assumptions through traditional user testing methods** (32:19) — Observing real users interact with software prevents product teams from fabricating new problems.

## Related Moments

- [Scaling AI adoption to non-traditional enterprise developers](https://www.wearedevelopers.com/videos/100256-can-this-elephant-dance-ibm-bob-and-the-future-of-ai-first-software-development) (from "Can This Elephant Dance? IBM Bob and the Future of AI-First Software Development")
- [Designing agentic AI solutions for the enterprise](https://www.wearedevelopers.com/videos/1831-ai-for-enterprise-developers-dr-damir-dobric) (from "AI for Enterprise Developers - Dr. Damir Dobric")
- [Exploring AI agent usage within the software engineering industry](https://www.wearedevelopers.com/videos/1814-wearedevelopers-live-markdown-liquid-and-checkouts) (from "WeAreDevelopers LIVE - Markdown, Liquid and Checkouts")
- [Deploying AI agents for enterprise legacy code modernization](https://www.wearedevelopers.com/videos/100256-can-this-elephant-dance-ibm-bob-and-the-future-of-ai-first-software-development) (from "Can This Elephant Dance? IBM Bob and the Future of AI-First Software Development")
- [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")
- [Shifting from AI hype to enterprise operations](https://www.wearedevelopers.com/videos/100328-the-limits-of-llms-in-real-world-applications) (from "The Limits of LLMs in Real-World Applications")

## Related Articles

- [Exploring AI: Opportunities and Risks for Developers](https://www.wearedevelopers.com/magazine/522-exploring-ai-opportunities-and-risks-for-developers)
- [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care)
- [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)

## Related Jobs

- [Principal Field Architect - AI Agents](https://www.wearedevelopers.com/jobs/ext/1442858-principal-field-architect-ai-agents) at **Twilio**
- [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**
- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Twilio's next Senior Principal Field Architect - AI Agents](https://www.wearedevelopers.com/jobs/ext/1487390-twilio-s-next-senior-principal-field-architect-ai-agents) at **Twilio**
- [Principal Product Manager, Agent Platform](https://www.wearedevelopers.com/jobs/ext/277541-principal-product-manager-agent-platform) at **GitHub**