Evidence Adejuwon @evidencecodes

Software engineer · founder @codetribeio · shipping production AI products · open to senior full-stack / founding eng roles at AI-first companies evidence.codetribehq.com Nigeria Joined May 2026

Tweets

39
Followers

48
Following

219
Likes

103

Evidence Adejuwon @evidencecodes

2 weeks ago

@swstica @dimensionalos evidence-codes

0 0 0 104 0

View Details

Evidence Adejuwon @evidencecodes

3 weeks ago

@SirAlexanderrr Very much interested

0 0 0 23 0

View Details

Evidence Adejuwon @evidencecodes

3 weeks ago

@quinn_leng That's a smart way to do it, let two months of even splits generate the real signal before committing to rules instead of guessing upfront.

0 0 1 5 0

View Details

Evidence Adejuwon @evidencecodes

3 weeks ago

@charliejhills The three-times-in-a-session trick is genius, way better than trying to predict every rule upfront. Letting Claude catch its own patterns instead of hand-writing rules for six months is such an obvious move in hindsight.

0 0 0 138 0

View Details

Evidence Adejuwon @evidencecodes

3 weeks ago

@weirdjacky This sounds exhausting, and it's clear you've tried everything. I'm sorry it's been this hard. Hope something opens up for you soon.

1 0 1 13 0

View Details

Evidence Adejuwon @evidencecodes

3 weeks ago

@danielatitienei Using chatgpt to translate the screenshots directly instead of just the text is a smart shortcut. Did you have to clean up any of the translations, or were they submit-ready?

0 0 0 19 0

View Details

Evidence Adejuwon @evidencecodes

3 weeks ago

@Fairywolf27 860 impressions to 13 downloads is a rough funnel but at least you have real numbers to work with. What was the thing it taught you that you didn't expect?

1 0 0 6 0

View Details

Evidence Adejuwon @evidencecodes

3 weeks ago

@BenjaminSc7519 Wait the $1.10 total AI spend against a self-enforced $25 ceiling is wild. How'd you land on one photo per tray instead of per card, still get clean reads at that scale?

1 0 1 26 0

View Details

Evidence Adejuwon @evidencecodes

4 weeks ago

@ZennenLeggett 310 lifetime downloads across three apps is a solid milestone, especially self driven traffic without a big following yet. Which of the three is pulling the most consistent daily numbers

1 0 1 14 0

View Details

Evidence Adejuwon @evidencecodes

4 weeks ago

@MattSwulinski @Superhuman @WisprFlow Wild that ten months of personal tooling turned into a funded direction. The bit about leverage only existing for people willing to live in a terminal is exactly why most internal tools never become products, curious to see how Viktor translates that.

0 0 0 36 0

View Details

Evidence Adejuwon @evidencecodes

4 weeks ago

Finished building Rekona today and got it installed on my own device, shipping it out soon. It reminds me 5, 3, 2, 1 days before a subscription charge hits so the card actually has funds by then, instead of another app that just helps you cancel things. Solves a very specific personal problem I had.

0 0 1 17 0

View Details

Evidence Adejuwon @evidencecodes

4 weeks ago

This is the real signal on AI native tooling. It's not about replacing Figma, Loom, or Trello, it's collapsing them into whatever shape actually matches how you think. Curious how handoff works now though, since the tool lives in one person's mental model. Do engineers adapt to it, or do you export back to something standard?

0 1 1 80 0

View Details

Evidence Adejuwon @evidencecodes

4 weeks ago

Most subscription apps are built to help you cancel things. My problem is the opposite. I want the charge to go through. I just forget to fund the card before it does, and then Claude declines mid-task or Netflix dies on a Friday night. So I built Rekona. It tells me 7, 5, 3, 2 and 1 day before every renewal, and on the day itself. Enough warning to actually move money. Not on the App Store yet. If this is your problem too, reply or DM me and I’ll add you to the first group testing it.

2 2 10 97 0

View Details

Evidence Adejuwon @evidencecodes

4 weeks ago

@FynCas MAKEUGC

0 0 0 744 0

View Details

Evidence Adejuwon @evidencecodes

4 weeks ago

@MichLieben REPO

1 0 0 26 0

View Details

Evidence Adejuwon @evidencecodes

a month ago

@everestchris6 GUIDE

0 0 0 20 0

View Details

Evidence Adejuwon @evidencecodes

a month ago

@lightsilver323 syntinel.codetribehq.com

1 0 2 170 0

View Details

Evidence Adejuwon @evidencecodes

a month ago

Really good point, and this is actually the direction I'm already leaning. The CLI is meant to support bringing your own model and provider, so you're not locked into one, and it also means nothing has to go through us at all if you point it at your own model. Likely open-sourcing the CLI itself for that reason. The sandbox execution and orchestration stays closed source though, that's the harder part to get right. Alongside all that, there's also a separate Semgrep and OWASP layer that doesn't depend on any model at all.

0 0 0 30 0

View Details

Evidence Adejuwon @evidencecodes

a month ago

How much of your code was written by AI lately? Mine too. Here's the problem nobody talks about: bugs that look fine, pass review, and ship anyway. So I built Syntinel. An AI that reads your codebase like an attacker would, SQL injection, hardcoded keys, auth bugs, prompt injection, then writes the actual exploit and the fix. One command. Real security audit. Not public yet, but here's a first look 👇 Comment "SYNTINEL" if you want in when this goes live.