Damboy retweeted
Honestly guys, we need to relax. It’s really not that hard.
If I fail, so what? I’ll try again. If I lose my job, so what? I’ll figure it out and try again.
I just have to keep showing up, keep taking my shots, and keep putting myself out there. Eventually, something is going to work.
We put so much unnecessary pressure on ourselves, worrying about things we haven’t even figured out yet.
The truth is, we’ll figure things out along the way. We always do.
So take a breath. Keep going. Enjoy the process.
Relax. It’s going to be okay.
Damboy retweeted
Over the last couple of months, my Team & I @Farmhub_agro have been building.
We have successfully built our first Tech Product, for our AgriTech Group.
It’s our Food Market App for our Food Market Company.
The way you subscribe for Data, Cable & other utilities. What can’t you subscribe for your Foodstuffs from the Market too?
You subscribe for your Foodstuffs, every week or every month so you don’t need to worry about that anymore. You just subscribe & we handle the rest.
We saw the challenges of the everyday people & we simply built a Solution.
The solution is simple.
All you need to do is Subscribe.
-We get the list of the foodstuffs you need. We handle the Market runs, which is directly from our own Food Market, where the food produce are hygienic and healthily grown.
-You don’t need to worry again about your Foodstuffs, we get it directly to you, whether it’s every week or every month,
based on your preference.
-The Icing, we have also integrated the Blockchain Technology where customers & users earn Loyalty points the more they use the App, the more the points they get and you can convert these points into Tokens to trade any currency in the future or use the points to get more foodstuffs or use these points at Hotels, Airports Lounges, other Supermarkets, etc.
It’s the Farmhub Food Market App & we’re just getting started!
Damboy retweeted
Wherever i am in the next 2-5 years, I pray I'm genuinely happy, successful, and living a life that once felt impossible.
Damboy retweeted
I have a lot of similar opportunities in my deals advisory pipeline vetted by myself and my team. If you’re a funds provider or just an HNI with liquidity open to financing deals of this nature, pls send me a DM
Pls RT
Damboy retweeted
I need you to start quantifying your efforts. It puts things in perspective.
1. Applying for a job: Don’t just apply to a lot of jobs daily, apply to 50 jobs daily.
2. Trying to save: Don’t just save everything, set a target.
3. Learning a skill: Set a timeline and milestones.
Damboy retweeted
get 6 months of free chatgpt pro + codex
OpenAI is recruiting Codex Ambassadors with perks including API credit
how to apply
1. go to developers.openai.com/commun…
2. record a 30 second intro video
3. share public links to projects you've built with Codex
4. submit before November 1
important
• not for students, almost for everyone
• you must publish content or organize an event every month
if you're already building with Codex, worth applying
1/13
A Zcash node could reject a bad block and still get stuck because of it.
Zellic found this in Zebra, Zcash’s Rust full node. The node rejected the block, but kept a record that later stopped it from processing the real one.
Here’s how that happened.
2/13
First, a detail about Zcash’s v5 transactions.
Transaction effects and authorizing data have separate commitments.
Changing some coinbase authorizing data leaves the transaction ID unchanged. Keep the original block header, and the block hash stays the same too.
3/13
That gave the researchers two different block bodies with the same block hash.
One was valid. The other failed validation.
The problem was that Zebra recorded the hash when sending the block for a write, and that record wasn’t properly cleared when the write failed.
4/13
When the valid body arrived later, Zebra found its hash in the cache and treated it as a duplicate.
The attacker had to get the bad body delivered first.
If they won that race, the leftover cache entry could stop the node from processing the valid block.
5/13
The attacker used the normal P2P flow: advertise a block hash, wait for Zebra to request it, then send the bad body.
In the PoC, sending the bad body first left the node at height 3.
Sending the valid body first let it reach height 4.
No RPC credentials were needed.
6/13
The fix was to clean up the bookkeeping when a block write failed.
Once Zebra rejected that body, the leftover record should no longer prevent it from trying the valid one.
This is why it’s worth following an error all the way through the cleanup code.
7/13
The second bug involved sync recovery.
A malicious peer replied to a request for block hashes with just two entries.
When Zebra requested one of them, the peer served a block whose coinbase height was far beyond the node’s current sync window.
8/13
Zebra’s downloader rejected that block before consensus verification.
But the resulting error hit a fallback that told the synchronizer to restart.
A bad response from one peer could therefore interrupt the node’s whole sync process. The peer could keep repeating it.
9/13
The local test made the impact clear.
The honest source was at height 120. Without the malicious peer, the victim also reached 120.
With the malicious peer, it finished at height 6.
The attacker had served just three of those high-height blocks during the test.
10/13
The fix changed how Zebra handled that error.
It no longer triggered a sync restart. Zebra also kept enough information about the advertiser to score the responsible peer.
Knowing which peer caused a failure matters when deciding how to recover from it.
11/13
Two smaller findings showed similar problems.
A response meant to contain at most 500 block hashes could pass through with up to 50,000.
And the inbound gossip path missed some peer misbehavior because it checked for the wrong error type.
12/13
My takeaway for audits: keep reading after the code rejects the input.
What did it add to a cache?
What work is still marked as pending?
Can an honest response succeed afterward?
Whose behavior triggers a retry or restart?
Then test whether the node actually recovers.
13/13
Zellic reports $93,750 in bounty awards across its Zcash research. The Zcash Foundation shipped fixes for the Zebra issues discussed here.
Full credit to Zellic for the research.
The original write-up is worth reading:
zellic.io/blog/zcash-protoco…
Getting @thallyio’s logo to show in Gmail sent me down a $990/year rabbit hole.
Got it working for $0. The missing piece was one Google profile setting.
Wrote up the steps for other solo and indie founders 👇 View article
Damboy retweeted
System Design is just DSA on another level of abstraction.
If you’re good at problem-solving, both LeetCode and System Design will be straightforward to you.
Damboy retweeted
timidan.xyz/lab
/lab ingests live info from my homelab
Damboy retweeted
"Alright folks let's rektoff"
One student put it that way in the chat, and week one of Cohort 6 went exactly like that.
The Discord took in a whole new cohort and already lives its own life. Intros, shared work, and questions often answered by other students first.
Meanwhile Jack Stodart ran two lectures:
> Lecture 1 started at the very bottom: how a Rust program lives in memory, from virtual memory and paging to the stack, heap and static memory
> Lecture 2 broke it on purpose. Memory errors in unsafe Rust, use-after-free among them, to watch undefined behavior with the guarantees off. After that, Rust's rules stop looking like red tape
Week one done, and the group is only getting warmed up.
Stay in safe Rust 🦀
Damboy retweeted
On @NewsCentralTV today for a conversation on my latest investigation, ‘Europe Through the Desert and the Mediterranean Sea’, I recalled how Arab traffickers stopped a Nigerian migrant from pausing to drink water after being forced to work as a manual labourer against his will.
I also spoke about 'Desserts from the Desert', my photo exhibition scheduled to be held at Alliance Française in Ikoyi, Lagos, next Monday, Tuesday and Wednesday.
Damboy retweeted
Cohort 6 is officially underway.
The kickoff town hall just wrapped, and we got to meet our newest Rektoffians. In this cohort, we have Rust backend engineers, Solidity devs moving over to Solana, and people from web2 bug bounty, joining from the US, Europe, and Asia.
They, in turn, met Jack Stodart, who leads Module 1. For the first three weeks, he takes them through how Rust handles memory: stack and heap, memory errors and unsafe code, then ownership. After that, he moves on to how the SVM handles memory and execution.
Lecture 1 is tomorrow. Until then, the group is working through the prep materials in Discord. Every lecture and office hour sits in one shared calendar, and every session is recorded.
Six weeks from now, the cohort finishes with a full capstone audit.
Time to lock in 🦀
Damboy retweeted
Schwartz-Zippel Lemma explains why evaluating polynomials at a single random point is sufficient to conclude that polynomial identity
u(x)v(x) = w(x) + h(x)t(x)
is true for all x, with high probability
Notes
github.com/t4sk/notes
Damboy retweeted
I pray life is kind to you as you help us reshare our business posts. 🤲🏾❤️
My husband is a professional driver, and we’re available for hires within Lagos for hangouts, outreaches, events, airport trips, offline rides and other programs.
Please RT and refer us to potential clients. Your support means a lot to our family. 🙏🏾❤️
































