
AI engineer building Web3 products and developer tools. 3ร hackathon winner ๐ Building @GholongHQ + @sport_402. Sharing build logs and lessons.
Life is a short trip โฐ๏ธ
Joined March 2020
- Tweets7.4K
- Following691
- Followers9.5K
- Likes10.7K
Building useful AI + Web3 products from Africa.
3ร hackathon winner. Building @GholongHQ and @sport_402.
I share build logs, demos, AI engineering lessons, Web3 payment infrastructure and product insights.
Follow the journey. More proof, less noise.
ALTCOIN D/G\DDY ๐น๐น๐ณ๐ฌ {A.D}๐ต retweeted
On-chain notice: bridge wallet rotation
As part of a wallet rotation, we have moved our multisig secured @chiliz bridge wallets into institutional custody. Major exchanges and @CoinMarketCap were informed ahead of the rotation.
These transfers are ours:
Chiliz Chain:
0xC94c4Ec62a59fCF4964873cC6529559Cc7717388 to 0xcd38983905eB4A433Fc44B3C90321522D8340BF2
Ethereum:
0x2F1d2754393356AEA6334180DA04BAB84412D580 to 0xcd38983905eB4A433Fc44B3C90321522D8340BF2
There is no change to circulating supply, no minting or burning of tokens, and no impact on CHZ trading.
A note from CHILIZยฎ CEO, Alex Dreyfus
On-chain notice: bridge wallet rotation
As part of a wallet rotation, we have moved our multisig secured @chiliz bridge wallets into institutional custody. Major exchanges and @CoinMarketCap were informed ahead of the rotation.
These transfers are ours:
Chiliz Chain:
0xC94c4Ec62a59fCF4964873cC6529559Cc7717388 to 0xcd38983905eB4A433Fc44B3C90321522D8340BF2
Ethereum:
0x2F1d2754393356AEA6334180DA04BAB84412D580 to 0xcd38983905eB4A433Fc44B3C90321522D8340BF2
There is no change to circulating supply, no minting or burning of tokens, and no impact on CHZ trading.
ALTCOIN D/G\DDY ๐น๐น๐ณ๐ฌ {A.D}๐ต retweeted
Grand rising!
XENO keeps building
Anticipate more update from the team ๐จ๐ฅ
ALTCOIN D/G\DDY ๐น๐น๐ณ๐ฌ {A.D}๐ต retweeted
5 months ago, I posted about launching an NFT collection that wasnโt just lore, but community-drivenโฆ, and for some reason fear and uncertainty crept in, and I couldnโt reach that goal, but I didnโt back off either.
I went back to my drawing board to be sure I really want to build this and since that 5months I have my training myself on what it takes to build up something with the help of hackathons.
It 5months down and yeah Iโm certain to push this but not alone., a solid team has being built, and you all my 7.9k family.
I thought about staying anonymousโฆ despite successful NFT projects I have worked on, but yeah, there is no harm in trying, and Iโm willing to see where this step takes my teammate and me.
So Xeno social is life and I canโt wait to dish out the lore behind Xeno but before then follow Xeno on X
@XENOonRobin and anticipate what my team and I are cooking up!
More info will roll out this week!
I will be in your DMs and even in your community.
ANTICIPATE๐จ
3 years of showing up and building in this space.
From memes to products, NFTs to communities. Iโve been in the trenches, learning, growing, and doing it with people I genuinely love.
And today, Iโm proud to announce Iโm launching my very first project.
This oneโs different. This one is a movement.
Structures are in place
The team is in place.
The branding is in motion. And the backend? Trust the process.
But I canโt do this alone and Iโm not trying to.
Iโm calling in communities, founders, and teams who want to move with us, build with us, and be part of something real from the ground up.
Socials coming soon.
More details dropping shortly.
Whatever the outcome Iโm just grateful I started.
Rooting for myself and my team.
Canโt wait for you all to meet XENO๐จ
62 franchises. $498B in combined value.
The NBA's average team is worth $6.64B, up 21% YoY.
The NFL's average team is worth $9.34B, up 31% YoY.
Yet direct ownership remains largely inaccessible to the broader public.
Would you consider owning equity in your favourite team?
A useful agent plan should answer four questions:
โข What outcome are we trying to reach?
โข What evidence would change the plan?
โข Which action needs approval?
โข When should the agent stop?
A plan that cannot change is just a longer prompt.
A plan helps only when the agent can revise it after new evidence. A useful design starts by asking what must be true before the system deserves more responsibility.
An AI memory feature needs four rules:
1. Save only information that improves a future task.
2. Attach a source and date.
3. Let the user inspect and correct it.
4. Expire it when it becomes unreliable.
Memory without control becomes confident clutter. #AIAgents
Useful agent memory is selective. Saving everything creates noise, privacy risk and stale assumptions. A useful design starts by asking what must be true before the system deserves more responsibility.
Which context failure causes the most trouble in your systems? Reply with the tradeoff you have seen in practice.
A practical context design checklist:
1. retrieve only useful context
2. label source and freshness
3. remove conflicting instructions
Test the smallest complete loop before expanding it.
Agent quality often depends more on relevant context than on a longer prompt. A useful design starts by asking what must be true before the system deserves more responsibility.
A practical tool selection checklist:
1. keep only necessary tools
2. describe each tool precisely
3. test wrong-tool selection
Test the smallest complete loop before expanding it.
More tools do not automatically make an agent more capable. Every tool adds another failure path. A useful design starts by asking what must be true before the system deserves more responsibility.
Where should an agent's responsibility end and human judgment begin? Reply with the tradeoff you have seen in practice.
A practical agent scope checklist:
1. define one outcome
2. limit the tools it may use
3. write a clear stop condition
Test the smallest complete loop before expanding it.









