Venture Builder Chronicles

Venture builder on a mission to create an ecosystem of profitable digital assets

Sharing raw experiments, lessons learned, and the journey along the way

[Venture Builder Chronicles @vb_chronicles]

Today I thought, what if there are two kinds of cognition.

Proactive: thought generates action. The cycle (thought, action, thought) builds momentum from the inside. Reactive: action generates thought. The cycle (action, thought, action) builds momentum from contact with the world.

Interesting part. For a reactive person, a thought can be the action, so the loop becomes thought, thought, thought. For a proactive person, action is already the thought, so it's action, action, action.

That said, the proactive mind wins by default in the game of performing actions. But the reactive mind isn't losing. It's just a different game. The reactive person shouldn't hunt for thoughts that spark action. The task is to train their RAS to hunt for actions that spark thought. Ideally ones already within reach.

[Venture Builder Chronicles @vb_chronicles]

Whether you practice 10 000 times or 1 000 times, everything will happen when the time comes, if you don't practice, that time will never come

[Venture Builder Chronicles @vb_chronicles]

Shipped a production prompt in 2 hours using my "Brainstorm-to-Spec" pipeline

The problem I was faced with is replying to client emails in my sourcing workflow eats time. Lots of it.

The output: a prompt, wired into the workflow.

Level 1 AI, nothing fancy. Saves a few minutes here, a few minutes there.

There are a million prompts like this online and this isn't even the first one I've used for email. But when you build it yourself, it feels differently.

Bigger thing I keep noticing:

the more automation I ship that actually works in a real workflow, the more I see simplicity wins.

[Venture Builder Chronicles @vb_chronicles]

@KodyTools That's true, and it sounds different when understanding comes from experience, instead of just because it's the logical way to think

[Venture Builder Chronicles @vb_chronicles]

What matters is what skills you want to build, what impact you want to make, and what you can actually do.

Labels, titles, descriptions, endless thinking about thinking, that's what has the least impact.

Reflections born from experience are very different from reflections born from thinking

[Venture Builder Chronicles @vb_chronicles]

Today learned about Context7. Basically it's a database of framework docs that helps llm's handle tasks better.

Looks useful, but I'm wondering: wouldn't it be better to just give for example Claude Code a direct link to the docs or connect a specific MCP server?

And, if you decide to work with that solution, wouldn't it be distracting for llm to be connected to the tool via mcp, as well as getting guidance from Context7?

[Venture Builder Chronicles @vb_chronicles]

Mastering Claude Code

Find out about spec driven workflow and started practicing it. The idea is you write a specification in non tech language, run it through proper deep technical planning, implement the plan, commit the results.

Created a spec template, commit message template, improved my understanding of plan mode.

Also liked the concept of constantly updating claude md. You rename your Claude md and run /init to get a fresh one based on your current project state.

[Venture Builder Chronicles @vb_chronicles]

Got interested in OpenClaw.

I actively use Claude but turns out that lately Anthropic doesn't allow using subscription OAuth tokens in third-party tools like OpenClaw.

Understandable.

So you need a separate provider for the lobster.

First thought:
Add Codex at $200 to subscriptions stack. OpenAI officially allows subscriptions in OpenClaw.

Reliable, but $400+/month on AI subscriptions with zero revenue at the current stage of my studio is a questionable investment.

Second thought:
OpenRouter. $50-100 in credits. For example, Kimi for routine tasks and monitoring, other models for other tasks, Claude API when you need heavy analysis.

Spent $100, stopped. Full control and flexibility.

Seems valid

Anyone tried this approach?

[Venture Builder Chronicles @vb_chronicles]

I used to do all my strategizing and ideation in ChatGPT.

Past few days since I started actively using Claude and Claude Code, my focus shifted.

Outcomes are better so far. And in the meantime Codex opened up for me from a completely different side.

Claude + Claude Code + Codex is a powerful combo

[Venture Builder Chronicles @vb_chronicles]

Agents are the new SaaS

Agree or Disagree?

[Venture Builder Chronicles @vb_chronicles]

Upwork is actually great for product research

Just open it, filter by $500+ jobs and look at what people are paying for right now.

You'd be surprised how many product ideas are just sitting there.

Just today in a matter of hour I found 3+ productized service ideas, and 1 product idea

But shipping is a different story xDDD

[Venture Builder Chronicles @vb_chronicles]

@stayopsteve Offline game is the next level game

[Venture Builder Chronicles @vb_chronicles]

Lately saw more posts where people mentioning directories as the new easy asset to build.

If the directory is genuinely useful, sure but when everyone is building the same thing, how many of them actually survive?

If you see this post, curious about your opinion

[Venture Builder Chronicles @vb_chronicles]

@TheGrowthLedger Documenting process :)

[Venture Builder Chronicles @vb_chronicles]

Back to posting!

Challenge is still going, now day 160, it changed a bit and got divided into chapters

Currently Chapter 1: $0 to $120

Today was all about refining the venture studio roadmap

Good day

[Venture Builder Chronicles @vb_chronicles]

@fortelabs Actually, all of them, it depends on the task

[Venture Builder Chronicles @vb_chronicles]

@brian_armstrong 100%

Diversification is good.

Having different types of assets works amazingly.

If you've been in stocks or savings, adding a small amount in crypto is the way to go.

[Venture Builder Chronicles @vb_chronicles]

Days 27-38 of "0 to 1 Challenge"
From Resistance to Building

I haven't posted a challenge update in a while.

Not because I quit, but because there were a lot of things to refine.

The Stall (Days 27-31)

Days 27-28: No challenge progress. Time was invested in other work-related activities.

Day 29: Still no progress, but I had a new framework idea. Yep, again xD.

The new system:

First 1-2 hours every morning - challenge work. No reading, no consuming content. If stuck, stare at the ceiling until building starts. Just no consumption.

Days 30-31: Weekend trip. Decided weekends are for rest, not forcing challenge work.

The Shift (Days 32-37)

Day 32: Watched a video where the presenter suggested starting with a Chrome extension because it's way easier than launching a web app. That resonated.

I changed my content consumption approach. Instead of watching entire tutorials, I only consumed the specific insight I needed: how to create a Chrome extension structure. Skipped intro, publishing, and debugging sections. Just the creation part.

Filled out my entire product canvas in a few hours. Refined my old process (hit some friction points, noted for later improvement, but not priority now).

Developed detailed PRD. Ready to code.

The Product:

A Chrome extension that automatically hides the "Play" button on Chesscom or Lichess after you lose 3 consecutive games in the past hour.

The Problem:
Chess players play on tilt. After 2-3 brutal losses, emotions take over. They keep clicking "Play Again" even though focus is shot.

One bad session = 8+ straight losses, 60+ point Elo drop, days of demoralization.

The Solution:
Hide Play button after 3 losses, 12-60 min cooldown timer, message: "You're tilted. Review, not play."

Target: Emotionally self-aware chess players (1000-1800 Elo) who know tilt is their real opponent and want systems over willpower.

Day 33: Technical setup, PRD refinement.

Day 34: Started with building session. Learned, or better to say, revoked my knowledge on how to use Git and GitHub.

Then developed first version of extension. It works in test mode!

Added donation feature (I think the product will be free to use), tested functionality.

The irony: 33 days gathering energy, then 80% of product built in one day.

Also updated my workflows.

Day 35: Built real-time loss streak detection. Tested extension functionality including DOM updates. Everything works great.

Day 36: Created PNG icons.
Added Buy Me a Coffee to popup after successful cooldown completion. Updated cooldown

Day 37: Rest day. Weekends for travel and recovery.

Day 38: Rest day. Weekends for travel and recovery.

What I learned?
What changed between Days 1-31 (constant research, no shipping) and Days 32-37 (actual building)?

Consumption filter:
Only consume exactly what I need for the current step, nothing more

Morning building ritual:
First hours are for building, not reading

Product choice:
Extension instead of web app as the first product during this run to decrease friction

The challenge isn't over.

After 33 days of resistance, building for 5 days straight felt natural. Not forced. Not obligatory.

Current status:

Extension functionally complete.
Icon and customization polish remaining.
Fix a few bugs and somewhere this week it will be launched.

Days until challenge ends: 82
Revenue: Still $0.00
Progress: Actually building something real

The documentation continues...

[Venture Builder Chronicles @vb_chronicles]

@trikcode And nowadays they are additionally becoming better at using ChatGPT, CodeRabbit and Claude

[Venture Builder Chronicles @vb_chronicles]

@agazdecki This is called HOPE