In this episode, Jerry discusses key insights on delegating to agentic tools while maintaining high levels of engineering ownership with Ozzie Osman, co-founder @ Monarch. Ozzie shares what it looks to pursue two paths when it comes to AI: agentic-forward and human-orchestrated pipelines. They also cover the specific AI tools that are used in Monarch, including Devin and Voltron; assigning tasks to be AI-first vs. human-led; determining which pieces of customer feedback lead to new features; and common challenges when it comes to AI-generated code and what the human review process for it looks like. 

ABOUT OZZIE OSMAN

Osman (Ozzie) Osman is co-founder of Monarch Money. He is the lead author of the Holloway Guide to Technical Recruiting and Hiring. He has built products and engineering teams at companies including Quora and Google. Ozzie has also started two companies that have been acquired, and advised dozens of other startups.

Sinch is the communications infrastructure the AI era runs on.

There's a layer of infrastructure behind every text, call, and login code your product sends, and it works exactly like plumbing: nobody thinks about it until it's the reason something broke.

Most providers route through 4-6 intermediaries; Sinch connects in 1-2 hops, direct carrier relationships across 600+ connections, handling 900 billion interactions a year across 60+ countries.

Routing, compliance, fraud prevention handled automatically rather than manually managed by the business sending the message!

Sinch is the reliability layer underneath AI-driven customer communications; the infrastructure that determines whether an AI agent's output actually reaches a real person as a delivered text, connected call, or verified interaction.

Check it out here!

SHOW NOTES:

• Introducing Ozzie & his role @ Monarch (2:37)
• Where Monarch is at in terms of technology transformation / user adoption (5:20)
• Insights on how Monarch is embracing AI with a security-forward mindset (7:21)
• What it looks like to pursue both agent- and human-orchestrated AI pipelines (11:00)
• Dissecting how an MCP shared infrastructure improves productivity (13:11)
• Inside the enablement team @ Monarch (15:25)
• Why Monarch uses Devin / prioritizing autonomy (16:33)
• How to fine-tune an agent to make it reusable (18:40)
• Strategies for determining what is an AI-centric vs. human-centric task (20:47)
• Understanding when & why Devin fails (24:49)
• Devin’s role in assisting the human-orchestrated processes (26:39)
• What a typical flow looks like throughout an SDLC (28:05)
• The importance of ownership in engineering when reviewing AI-generated code (29:31)
• Frameworks for prioritizing feature requests based on user feedback (35:15)
• Ozzie’s perspective on the role of engineering teams (40:03)
• Final thoughts on navigating AI-related anxieties (43:08)

This episode wouldn’t have been possible without the help of our incredible production team:

Patrick Gallagher - Producer & Co-Host

Jerry Li - Co-Host

Noah Olberding - Associate Producer, Audio & Video Editor https://www.linkedin.com/in/noah-olberding/

Dan Overheim - Audio Engineer, Dan’s also an avid 3D printer - https://www.bnd3d.com/

Ellie Coggins Angus - Copywriter, Check out her other work at https://elliecoggins.com/about/