How to Build AI Agents That Check Their Own Work | Jared Zoneraich

How to Build AI Agents That Check Their Own Work | Jared Zoneraich

Author: Peter Yang July 12, 2026 Duration: 47:12

Jared is a Builder in Residence at Cognition, the team behind Devin. He showed me why teams often overbuild their first agents, and how simple prompts, better tools, and self-testing work together instead. Then he demoed one manager agent breaking down a task, directing 10 cloud agents, and deciding what a human needs to review.


Jared and I talked about:

(00:00) Why teams overbuild their first AI agent

(02:50) Let the model cook before adding rules

(03:16) Why tools matter more than better prompts

(10:51) Why you should build evals after shipping

(16:06) What cloud agents can do that local agents can’t

(20:30) How to get one agent to manage other agents

(25:39) Demo: One agent launches a team of 10

(28:03) How agents prove their work before you review it

(33:18) The best tasks for agent teams to work on


Thanks to our sponsors:

Riverside: All-in-one AI studio for podcasts and video https://creators.riverside.com/PeterYang

Linear: The AI agent platform for modern teams https://linear.app/behind-the-craft

Wispr Flow: 4x faster than typing with your voice https://ref.wisprflow.ai/peteryang


📌 Get the takeaways: https://creatoreconomy.so/p/how-to-build-ai-agents-that-check-their-own-work-jared-cognition-devin


📌 Get my personal AI OS with useful AI skills, $270 in AI tool credits, live monthly workshops, and courses on Hermes and Codex agents: https://www.behindthecraft.com/


Where to find Jared:


X: https://x.com/imjaredz

Devin: https://devin.ai


Subscribe to this channel - more interviews coming soon!


For anyone building the future, Behind the Craft is a conversation with Peter Yang that moves beyond theory and into the tangible details of creation. This podcast lives in the messy, rewarding space where ideas become real products. Each episode is built on candid interviews with experts who have been in the trenches, dissecting the pivotal decisions, the unexpected hurdles, and the hard-won lessons that rarely make it into a polished case study. You’ll hear the unvarnished stories behind the features and companies shaping our world, focusing on the practical frameworks and mental models that effective product leaders and creators rely on daily. It’s about understanding the craft from the inside out-the strategic shifts, the team dynamics, the user insights that truly move the needle. Tune in for a direct, no-fluff dialogue designed to accelerate your own journey, providing actionable guidance you can apply immediately to level up your own work. This is where the blueprint meets the build.
Author: Language: English Episodes: 50

Behind the Craft
Podcast Episodes
Master Google AI Studio in 40 Minutes | Logan Kilpatrick [not-audio_url] [/not-audio_url]

Duration: 39:17
Logan is the Product Lead for Google AI Studio. I got him to give us an inside look at how he uses AI Studio to build AI Studio (very meta) and how his team ships at startup speed inside Google. AI Studio is the feature…
Build an AI Analyst with Claude Code in 50 Min | Sumeet Marwaha [not-audio_url] [/not-audio_url]

Duration: 51:47
Sumeet is the Head of Data at Brex and came personally recommended by the Claude Code team. In our episode, he showed me how to use Claude Code to build a data explorer that lets anyone ask questions and get insights wit…
A Founder's Playbook for Shipping 10x Faster with AI | Yana Welinder [not-audio_url] [/not-audio_url]

Duration: 41:53
Yana is Head of AI at Amplitude and a good friend. We had some real talk about how to stay scrappy inside a big company, including how to avoid decision by committee and endless internal debates. Yana also demoed her fav…
A Proven 5-Step System to Prototype Apps with AI | Xinran Ma [not-audio_url] [/not-audio_url]

Duration: 38:14
Xinran is a top AI instructor who has taught 100s of product teams how to design and prototype with AI. He shared 5 practical techniques to avoid generating AI slop — from side-by-side exploration to reverse prompting to…