Lenny's Podcast: Product | Career | Growth

chevron_right

How Block is becoming the most AI-native enterprise in the world | Dhanji R. Prasanna

whatshot 1676 snips

Oct 26, 2025

Guest

Dhanji R. Prasanna

Dhanji R. Prasanna, Chief Technology Officer at Block, has managed a massive engineering team driving AI innovation. He reveals how Block’s open-source agent, Goose, saves employees significant time weekly. Dhanji discusses the surprising productivity gains for non-technical teams and the importance of leadership in adopting AI tools. He shares insights on code quality versus product success, the impact of organizational changes, and lessons learned from previous product failures. Dhanji emphasizes starting small with projects to foster innovation.

01:26:41

forum

Ask episode

web_stories

AI Snips

view_agenda

Chapters

menu_book

Books

auto_awesome

Transcript

info_circle

Episode notes

Enable Non-Technical Self-Service

Empower non-technical teams to build their own tools with agents instead of waiting for centralized engineering.
Encourage teams to self-serve using Goose to compress weeks of work into hours for internal workflows.

Rewriting Over Refactoring

Future engineering will shift from incremental refactors to frequent full rewrites as agents can rebuild apps overnight.
Dhanji imagines routinely throwing away and regenerating apps to explore many experiments in parallel.

Build Only When It Serves Your Mission

Focus on your company's core purpose when deciding to build versus buy; avoid chasing pure cost savings.
Don't lose product focus and mental bandwidth by rebuilding vendor functionality unless it serves your mission.

Get the Snipd Podcast app to discover more snips from this episode

How Dhanji Convincing Leadership of AI

06:01 • 2min

chevron_right

Reorganizing Block to Be Technology-First

07:54 • 4min

chevron_right

Conway's Law and Team Collaboraion Changes

12:14 • 2min

chevron_right

How Engineers Work Differently with AI

14:13 • 1min

chevron_right

Measuring AI Productivity Gains

15:30 • 2min

chevron_right

Non-Technical Teams Building Tools with AI

17:57 • 1min

chevron_right

Automating Mobile UI Tests with Gosling

18:59 • 3min

chevron_right

What Goose Is and the Model Context Protocol

21:50 • 2min

chevron_right

Goose: Open Source, Extensible Agent Platform

23:55 • 3min

chevron_right

Why Block Open-Sourced Goose

26:44 • 2min

chevron_right

An Engineer Letting Goose 'Watch' His Work

28:50 • 4min

chevron_right

Future Workflows: Autonomous, Always-On Agents

32:32 • 4min

chevron_right

AI Successes and Where Humans Still Lead

36:40 • 2min

chevron_right

When Teams Should Question Building New Tools

38:31 • 2min

chevron_right

Build vs. Buy with AI in Mind

40:30 • 3min

chevron_right

How AI Affects Hiring and Roles

43:13 • 4min

chevron_right

Interviewing with AI Tools

47:33 • 1min

chevron_right

Who Benefits Most from AI Tools

48:58 • 3min

chevron_right

Organizational Changes Often Outperform Tools

52:22 • 2min

chevron_right

Drive Adoption by Using Tools as Leaders

54:04 • 2min

chevron_right

Using Goose for Personal Productivity Tasks

55:42 • 2min

chevron_right

How Anyone Can Try Goose

58:09 • 2min

chevron_right

What Dhanji Would Have Told His Earlier Self

01:00:03 • 2min

chevron_right

Counterintuitive Lesson: Code Quality vs. Product Success

01:02:02 • 7min

chevron_right

Start Small: Hackweeks to Big Products

01:09:24 • 5min

chevron_right

Learning from Repeated Product Failures

01:14:09 • 2min

chevron_right

Leadership Lessons: Start Small and Question Assumptions

01:16:00 • 2min

chevron_right

Books, Shows, and Personal Picks

01:18:12 • 6min

chevron_right

How to Find Dhanji and Contribute

Alfred, Lord Tennyson

This volume includes a substantial selection of Alfred Tennyson's poetry, featuring complete poems such as 'The Lady of Shalott,' 'The Lotos-Eaters,' 'Ulysses,' 'Break, Break, Break,' and 'The Charge of the Light Brigade,' alongside selections from longer works like 'The Princess,' 'In Memoriam,' and 'Maud. ' The collection is often edited with contextual notes and a chronology of Tennyson's life and times, providing insight into his poetic development and Victorian context. It is considered a definitive introduction to Tennyson's poetry and his status as one of the greatest poets of the Victorian era.

#1514

• Mentioned in 27 episodes

The Master and Margarita

Mikhail Bulgakov

The novel intertwines several narratives, including the story of Pontius Pilate and Jesus Christ, and the adventures of the devil Woland and his entourage in Moscow. The Master, a writer who has been driven to insanity by the rejection of his novel about Pontius Pilate, and his mistress Margarita, who remains loyal and determined to save him, are central characters. Margarita's journey involves becoming a witch and hosting Woland's midnight ball, where she uses her newfound powers for altruistic ends. The book is a critique of Soviet society, exposing corruption, moral decay, and the repression of artists and dissidents. It blends elements of satire, fantasy, and philosophical inquiry, reflecting Bulgakov's personal experiences with censorship and societal pressures.

Dhanji R. Prasanna is the chief technology officer at Block (formerly Square), where he’s managed more than 4,000 engineers over the past two years. Under his leadership, Block has become one of the most AI-native large companies in the world. Before becoming CTO, Dhanji wrote an “AI manifesto” to CEO Jack Dorsey that sparked a company-wide transformation (and his promotion to CTO).

We discuss:

1. How Block’s internal open-source agent, called Goose, is saving employees 8 to 10 hours weekly

2. How the company measures AI productivity gains across technical and non-technical teams

3. Which teams are benefiting most from AI (it’s not engineering)

4. The boring organizational change that boosted productivity even more than AI tools

5. Why code quality has almost nothing to do with product success

6. How to drive AI adoption throughout an organization (hint: leadership needs to use the tools daily)

7. Lessons from building Google Wave, Google+, and other failed products

—

Brought to you by:

Sinch—Build messaging, email, and calling into your product: https://sinch.com/lenny

Figma Make—A prompt-to-code tool for making ideas real: https://www.figma.com/lenny/

Persona—A global leader in digital identity verification: https://withpersona.com/lenny

—

Where to find Dhanji R. Prasanna:

• LinkedIn: https://www.linkedin.com/in/dhanji/

—

Where to find Lenny:

• Newsletter: https://www.lennysnewsletter.com

• X: https://twitter.com/lennysan

• LinkedIn: https://www.linkedin.com/in/lennyrachitsky/

—

In this episode, we cover:

(00:00) Introduction to Dhanji

(05:26) The AI manifesto: convincing Jack Dorsey

(07:33) Transforming into a more AI-native company

(12:05) How engineering teams work differently today

(15:24) Goose: Block’s open-source AI agent

(20:18) Measuring AI productivity gains across teams

(21:38) What Goose is and how it works

(32:15) The future of AI in engineering and productivity

(37:42) The importance of human taste

(40:10) Building vs. buying software

(44:08) How AI is changing hiring and team structure

(53:45) The importance of using AI tools yourself before deploying them

(55:13) How Goose helped solve a personal problem with receipts

(58:01) What makes Goose unique

(59:57) What Dhanji wishes he knew before becoming CTO

(01:01:49) Counterintuitive lessons in product development

(01:04:56) Why controlled chaos can be good for engineering teams

(01:08:07) Core leadership lessons

(01:13:36) Failure corner

(01:15:50) Lightning round and final thoughts

—

Referenced:

• Jack Dorsey on X: https://x.com/jack

• Block: https://block.xyz/

• Square: https://squareup.com/

• Cash App: https://cash.app/

• What is Conway’s Law?: https://www.microsoft.com/en-us/microsoft-365-life-hacks/organization/what-is-conways-law#

• Goose: https://github.com/block/goose

• Gosling: https://github.com/block/goose-mobile

• Salesforce: https://www.salesforce.com/

• Snowflake: https://www.snowflake.com/

• Claude: https://claude.ai/

• Anthropic co-founder on quitting OpenAI, AGI predictions, $100M talent wars, 20% unemployment, and the nightmare scenarios keeping him up at night | Ben Mann: https://www.lennysnewsletter.com/p/anthropic-co-founder-benjamin-mann

• OpenAI: https://openai.com/

• OpenAI’s CPO on how AI changes must-have skills, moats, coding, startup playbooks, more | Kevin Weil (CPO at OpenAI, ex-Instagram, Twitter): https://www.lennysnewsletter.com/p/kevin-weil-open-ai

• Llama: https://www.llama.com/

• Cursor: https://cursor.com/

• The rise of Cursor: The $300M ARR AI tool that engineers can’t stop using | Michael Truell (co-founder and CEO): https://www.lennysnewsletter.com/p/the-rise-of-cursor-michael-truell

• Top Gun: https://www.imdb.com/title/tt0092099/

• Lenny’s vibe-coded Lovable app: https://gdoc-images-grab.lovable.app/

• Afterpay: https://github.com/afterpay

• Bitkey: https://bitkey.world/

• Proto: https://github.com/proto-at-block

• Brad Axen on LinkedIn: https://www.linkedin.com/in/bradleyaxen/

• Databricks: https://www.databricks.com/

• Carl Sagan’s quote: https://www.goodreads.com/quotes/32952-if-you-wish-to-make-an-apple-pie-from-scratch

• Google Wave: https://en.wikipedia.org/wiki/Google_Wave

• Google Video: https://en.wikipedia.org/wiki/Google_Video

• Secret: https://en.wikipedia.org/wiki/Secret_(app)

• Alien Earth on FX: https://www.fxnetworks.com/shows/alien-earth