Semi profitable trader
Joined April 2022
- Tweets2.2K
- Following2.8K
- Followers6.4K
- Likes10.7K
Bro so deeply underwater on his Pons short he said fk it let me just try anything 😭
This quoted post is unavailable.
OGs were also genuinely trading against North Korea during defi summer with a fraction of the tools btw
jbjbjb retweeted
savUSD is now live on @Movement_xyz.
Institutional-grade yield infrastructure, natively on Movement.
Deposit stablecoins, earn yield automatically, protected by the same multi-layered capital stack that's had zero incidents since launch.
Anthropic cutting OpenClaw is a massive net benefit imo.
Instead of “one model to rule them all” this pushes users to “route to the best model for specific tasks”
People have been paying the most expensive model for every task. The next phase for AI will be “smart routing”
Although 95% of what happened with Drift today was gross negligence by the team, @circle has a duty and a responsibility to not facilitate native bridging of stolen funds:
• 3+ hours after the exploit was publicly flagged
• During normal US business hours
• While the attacker was actively using CCTP
• One week after Circle froze 16 innocent business wallets within hours over a civil lawsuit filled w errors
Why does Circle not have a full time response team dedicated for events like this?
Do better.
The USR incident wasn't a "hardcoded peg" failure or even a vault problem.
It was a curator conflict of interest amplified by an automated allocator with no off switch.
Morpho's isolated markets work fine. The MetaMorpho vault layer is a UX wrapper. The wrapper failed, not the protocol.
What actually needs to change:
1. Automated allocators need circuit breakers. The public allocator kept feeding USDC into a broken market post-exploit. Most of the bad debt was created after the initial hack.
2. Curators managing proprietary funds + public vaults simultaneously is a massive conflict of interest. Who gets to exit first in stress situations?
3. Hardcoded oracles on mutable-supply assets need guardrails on the mint side, not the oracle side.
I think hardcoded redemption oracles are actually the best move for most defi stables.
Lending markets live on stable leverage. Protecting those users from flash depeg wicks via illiquid on chain LPs is more important for overall defi stability than most people realize.
The future is permissioned markets with proper access control. Not removing the oracle design that actually protects leveraged positions from unnecessary losses.
Most memory tools just do vector search.
SME uses 6 scoring signals on every query:
1. FTS5 keyword matching (instant, no embeddings needed)
2. Temporal awareness ("what did I decide last week?" just works)
3. Confidence decay (old irrelevant memories fade automatically)
4. Entity graph overlap (ask about "Sarah" → also get her project context)
5. Type-aware scoring (decisions > random notes)
6. File weight (your important files rank higher)
The result: the right memory surfaces first. Every time
Your AI agents forget things, this fixes that.
Works with Claude Code, Codex, OpenClaw, Cursor, or any agent that reads context.
in terminal, run: npm install structured-memory-engine
github.com/Bryptobricks/Stru…
1,451 tests. 0 failures. Built in production. Open source.
Coinbase, Robinhood, Binance. No amount of exchange listings will save you from terrible tokenomics.
LLP now being entirely dependent on $LIT price action is incredibly risky to the system's health and overall performance. Hyperliquid killer tho.
