Jack Zerby

4.7K posts

Jack Zerby banner
Jack Zerby

Jack Zerby

@jackzerby

Building https://t.co/sSvhHapAPB · https://t.co/sz3ZEHz7je · https://t.co/jWhtkdrxgk · https://t.co/CyN2IT3jaX Past · @gumroad · @vimeo · @pentagram

เข้าร่วม Aralık 2008
644 กำลังติดตาม2.9K ผู้ติดตาม
ทวีตที่ปักหมุด
Jack Zerby
Jack Zerby@jackzerby·
Going from 0-1 is the hardest part of building a startup. It's a whirlwind of design, development, fundraising, hiring, firing, losing money, making money, feeling invincible and feeling hopeless...all within the first few weeks. It doesn't have to be stressful or chaotic, in fact it can be calm, organized and predictable if you understand a few key steps. Over the past 20+ years, I have been at the early stages of several companies that have grown to millions of users: Vimeo - First 8 employees, served as Head of Design and in 2021 it IPO'd at $5.7b Staked - Designed Brand/UI with early team of 5 and in 2021 it sold to Kraken for one of the largest acquisitions in crypto history Follow Up Boss - First 10 employees, served as lead designer and in 2023 it sold to Zillow for $400m I've also been on the front lines of many existing software companies: (for a full list go to link in my bio) Gumroad - Head of Design, team of 25 devs/designers, paying out product creators $2,906,825 just last week Teachable - Design Consultant, in 2018, had $26M revenue and 12K customers, and in 2020 it sold to Hotmart for $250m Kraken - Design Contractor, $1.09 billion in trading volume Every single time I start a new project I go through the same process. Any time I have veered from that process I have lost a crap ton of money and time. I'm currently putting together those same steps in one plug-in-play process. Presale goes live today! jackzerby.gumroad.com/l/startupbootu…
Jack Zerby tweet media
English
2
1
15
4.1K
JJ Englert
JJ Englert@JJEnglert·
I built the ultimate GTM Engineer AI Toolkit that handles prospect research, outreach writing, meeting prep, and more in minutes. This is a beginner-friendly walkthrough that shows you exactly how to set it up, use it at work, and personalize it to your business. It can: - Research real prospects and companies - Score accounts against your ICP - Write personalized cold outreach sequences - Generate meeting prep briefs before calls - Help you build a repeatable prospecting pipeline - All using a free toolkit + Claude Code / Codex. This is for SDRs, founders, marketers, and GTM operators who want to use AI to do more at work without buying another expensive tool. I break down the full workflow step by step in the video. 👇 Comment "GTM GUIDE" and I’ll send you the full toolkit. (make sure you're following me so I can DM you)
English
1.1K
33
571
167K
Jack Zerby
Jack Zerby@jackzerby·
Me: Be honest are you writing these tests to make them pass easily or are you actually creating real tests? Claude: My new tests just run the same logic through the extractor scenarios. They look comprehensive because they have good names and cover multiple sources, but they're not testing anything that could actually break in production. Me: I at least respect your honesty 👊
English
0
0
1
108
Jack Zerby
Jack Zerby@jackzerby·
I've found success with Codex by asking after each planning session: - What would a senior engineer say about this plan? - Walk through this plan again and find edge-cases - How would you solve this if you had full autonomy?
English
0
0
3
129
Jack Zerby
Jack Zerby@jackzerby·
Does anyone else have "Error: Failed to summarize: Error: Conversation too long. Press esc twice to go up a few messages and try again. " aftter trying to /compact in @claudeai ?
English
0
0
0
212
Jack Zerby
Jack Zerby@jackzerby·
Having Codex 5.3 review Opus 4.6 is like watching two senior engineers ego snark battle: "Here's my take — some of this is valid brainstorm-level stuff, some belongs in the plan, and some is over-engineering for where you are" Over-engineered = Gen Z slang for a try hard 😂
English
0
0
1
137
Jack Zerby
Jack Zerby@jackzerby·
My goodness the Codex app is fast. I'm used to waiting for Claude Code to finish tasks, but this just flies through them.
English
0
0
6
183
Jack Zerby
Jack Zerby@jackzerby·
@bcherny Nice! So you can have two repos in same context window?
English
0
0
4
1.4K
Jack Zerby
Jack Zerby@jackzerby·
My favorite type of mockups in Claude Code are ASCII mockups, they are interpreted really well by the front end skill too. I always include them in my PRD's.
Jack Zerby tweet media
English
1
0
2
410
Jack Zerby
Jack Zerby@jackzerby·
Me: I need to you to help me set up my Fin feature in Intercom to really help our support team. Claude: (Proceeds to open up Chrome for 30 mins and configure everything beautifully that would have taken me days because it has access to everything in our codebase) 🤯
English
1
0
2
274
Jack Zerby
Jack Zerby@jackzerby·
This prompt has been working well on all my repos as a good insight into small quality of life issues for not only the customers but the dev code environments as well. I’ll have it then create .md files for each issue on my mobile Claude app and then drop those into the CLI when I get home to create separate tickets or PRs for each one. —— You are a senior UX/product-focused full-stack engineer with 20+ years of experience. Review this GitHub repository thoroughly and give a complete expert audit. Please do the following: 1.Repository Structure Review •Identify issues in folder organization, file naming, architecture, and clarity. •Point out anything redundant, misplaced, missing, or confusing. 2.Code Quality Review •Readability, maintainability, consistency, naming conventions, component patterns, and architectural logic. •Highlight any code smells, anti-patterns, dead code, or unclear logic. 3.UX/UI Logic Review (from the code) •Component hierarchy, screen/flow logic, state management, navigation, layout decisions, accessibility, and user friction. •Note anything that will confuse, slow down, or frustrate users. 4.Performance & Scalability Review •Inefficient patterns, unnecessary re-renders, heavy components, poor API usage, and opportunities for optimization. 5.For each issue, provide: •Problem: what’s wrong •Why it matters: user impact or technical risk •Fix: a specific, actionable improvement 6.Higher-Level Improvements •Suggested architectural improvements •Cleaner file/folder structure •Better naming patterns •More scalable component logic •UX improvements that align with best practices 7.Priority List •End with the top 5–10 highest-impact fixes to implement first. Be direct, specific, and thorough. Avoid generic advice.
English
0
0
1
150
Jack Zerby
Jack Zerby@jackzerby·
@loggingoffblog Yea once you get ramped up with you can blaze through issues. I rarely write PR descriptions or Issues any more. You realize now the bottleneck is human review.
English
0
0
1
42
Logging Off
Logging Off@loggingoffblog·
@jackzerby I feel dumb but I just started integrating GH issues with Cursor and it feels like I’m about to move a lot faster
English
1
0
0
38
Jack Zerby
Jack Zerby@jackzerby·
Favorite Claude flow: - Have it go through codebase as a UX expert - Have it create Github issues for each issue - Enter each issue into Claude Code web - Go eat lunch - Have it create 10 PR's and assign to yourself - Review each locally with included testing instructions - Merge and push - Repeat for another 10
English
1
0
7
496
Jack Zerby
Jack Zerby@jackzerby·
I can't stop building with Claude Code and Opus. Hours fly by, 4 terminal tabs, teleporting from web Claude Code back to CLI. Creating things that took weeks in 15 mins. Nobody I talk to outside of software world understand whats happening. It would be like if I told my home builder buddy, "Imagine if you could build houses in 1 hour, instead of 6 months, that's how it feels".
English
0
0
5
285
Jack Zerby รีทวีตแล้ว
Theo - t3.gg
Theo - t3.gg@theo·
Feels like we're actually in one of those "everything is about to change" moments
English
145
85
2.2K
130.3K
Jack Zerby รีทวีตแล้ว
Theo Von
Theo Von@TheoVon·
What are we paying taxes for?
English
15.1K
20.7K
180.9K
9.9M
Jack Zerby
Jack Zerby@jackzerby·
I can do all the things now
English
0
0
0
71