dhadrien (@dhadrien_) header
dhadrien (@dhadrien_) profile picture

orchestrating magics in my aivory tower, project to be revealed soon founded @sismo_eth zk in 2021, @variabl_io defi in 2016

Paris, France
Joined March 2014
Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)Photo by dhadrien (@dhadrien_)
We’ve just raised $175M, in a round co-led by @paradigm, @a16zcrypto, and @RibbitCapital, the largest raise DeFi has ever seen. Credit is the bedrock of our civilization, but the infrastructure underneath is fragmented, extractive, and closed to most of the world. That is what @Morpho is here to change. Morpho is building the open credit network for the world. The global credit market is $ 200T. We are building the infrastructure layer that will move it onchain, and every institution, fintech, and bank that wants to participate in the next era of finance will connect to this network. After four years of being heads down building Morpho, we now count more than $11B in deposits and integrations with leading financial institutions including @Coinbase, @Binance, @FireblocksHQ, @SG_Forge, @krakenfx, @Bitwise, and dozens of others building on Morpho to offer better products to their users. But this is just the beginning. This raise will allow us to accelerate: activating the global credit network at scale, building the go-to-market engine to match our ambition, and bringing more institutions onto the network faster. Having any kind of onchain strategy in this era will mean going through Morpho at some point. Reach out and let's talk. And if you want to help build the open credit network for the world, we have plenty of open roles. Let’s fly 🦋
Morpho Association has raised $175M to build the open credit network for the world. Co-led by @paradigm, @a16zcrypto, @RibbitCapital with strategic participation from @apolloglobal, @vaneck_us, @circle_ventures, and @Ledger @Cathayinnov. The round also included participation from @variantfund, @wmt_ventures, @preludexyz, @IOSGVC, @HashKey_Capital, @sbigroup, @Bpifrance, @mirana, @bamazizimesh, NJJ Capital and 10+ other strategic partners. The funding will help accelerate Morpho's position as the foundation for onchain credit.
331
140
55
2,372
396,334
stopped posting causing i’ve thrown my harness away, and learning phase quite over, was a great experiment to learn fast and low level about ai, would 100% recommend anyone to do so back to build my new project with all learnings, locked in, loving it, grateful for everything 🙏
3
10
264
when singing it’s so obvious we are predicting machines not so far from llms when you listen to a song, your mind is always half a sec in advance, predicting the key word, note, drum beat. even when it’s a first time listen
4
7
301
like llms we can’t help but to imagine the next note. like llms we eval ourselves and fine tune us and after 10 listens we predict better but not yet 100% accurate after 1000 listens we start to be accurate
1
178
spent last 3 months building harnesses, using openclaw codex claude cursor pi tuis & sdks. open weights w/ comfy ui and llama.ccp. would def recommend to anyone new to ai if you want to do more than understanding agents but really be in flow with them.
You get good at anything by doing it a lot. Just do things a lot It's not time spent. It's volume of doing. Make sure the time it takes you to do a single thing is as fast as possible. Your cycle time is sacred
2
3
379
you’ll cover: - spec driven vs iterative approaches for complex systems that harnesses - agent loop and observability of agents - how to test a hybrid system (agents + deterministic code) - how to manage an agent context - what models are good for what - know multiple ways of doing memory, context injection, compaction etc.. - how to make it so agent learn automatically - so many more things should write an article tbh
1
91
exactly, use frontends as review tools. claude is insanely good at frontends.
3
228
I think @antirez ds4.c is important! I wrote down my thoughts on why I built pi-ds4 and why we need to focus our local model efforts stronger than we do currently. lucumr.pocoo.org/2026/5/8/lo…
15
50
7
370
30,915
using pi for backend agentic workflows, its really good
1
71
created a meme generator tool for fun this morning asked him to meme me ty ai
2
182
taking a day of is chilling with only one gpt 5.5 fast mode in codex + reading bookmarked articles on x
78
totally agree, I think tomorrow's cofounders will be just that: one tech deep system expert one artist the marketings/ sales/ management will be about expanding the artist vibe via the system
Claude Code eng leader @Nerdi_Yogi on the two profiles she's hiring for now: 1. Creative builders with product sense 2. Deep systems experts (for the hard parts)
1
231
btw the goals of such a frontend is not usability, it's reviewability of the underlying system makes it 10x better to review your backend/ system with a frontend that surfaces all its capabilities (screenshots of the review frontend of a tool I built in < 2 hours)
workflow that works well when vibe coding tools: - build the backend with codex (workflows, backend, db etc..) - /loop in claude code: "Every 30 min, rebase from main and rebuild the frontend to surface everything from the system"
1
217
leveraging the most impressive capability of claude-code/ opus 4.7 btw which is building frontend on top of complex systems imo
73
workflow that works well when vibe coding tools: - build the backend with codex (workflows, backend, db etc..) - /loop in claude code: "Every 30 min, rebase from main and rebuild the frontend to surface everything from the system"
1
1
341
pi is great qwen uncensored is great (use for testing/ evals/ non demanding task + censored tasks like download stuff or analyse stuff)
156
codex / claude code to dev (make overall better dev agents) pi agent for in-app agents, specific harness (better observability, hooks etc..) am I wrong?
122
why not more people do post training on open weights like qwen? feel like not too many do?
86
stack for orchestration & harness: - claude code + codex sdks (or pi sdk): agent loops + compaction - shared MCP: read/write memories/skills/files creating agents and cross agents coms - event based gateway: crons, evals, hearbeats, pipelines, hooks etc.
203
claude-code vs codex: they just work so well together 1/ system-wide feature - opus 4.7 1m context to get systemic view, make it work on a plan - codex 5.5, will have a more pragmatic plan both ship their plan and push have them compare each others, they will collaborate 2/
2
2
318
2/ wide fix using /goal - make a /goal (like refactor, distill docs, modularize code etc..) - have opus 4.7 review against system every houre 3/ new a simple front + back + deb feature - codex 5.5 ship it first: all - opus 4.7 review, add tests, docs, spec then ship the front
1
122
generally claude 10x better on front claude 5x better on systemic thinking and following systemic rules codex 5x faster codex 10x better at incremental features/fixes
94