Pinned Tweet
How Relational Databases Work. This post talks about how indexes and transactions work on the inside of relational databases. architecturenotes.co/things-…
Architecture Notes retweeted
All these Rube Goldberg machines around the models are actually making them way worse.
This is both in products and workflows.
Architecture Notes retweeted
ssh ssh.mahdiyusuf.com
check it out.
Architecture Notes retweeted
At what point do we stop calling them javascript vulnerabilities and start calling javascript itself a feature-rich remote code execution platform.
Architecture Notes retweeted
The Speed Tax.
mahdiyusuf.com/the-speed-tax…
As many of you know every week we send out newsletter and each one has a new image the top each week, this takes a bit of time to design and create with @LumaLabsAI. I am about to produce this in the fraction of the time with tons of artistic control.
lumalabs.ai/uni-1
Uni-1 feels like a shift from prompting to actually collaborating.
Multimodal + native pixel generation opens up a lot creatively and allows for fast iteration and storytelling accurately.
Curious what people build 👀 lumalabs.ai/uni-1
Architecture Notes retweeted
America outsourced manufacturing, then the skills to design what gets manufactured. Software is next.
mahdiyusuf.com/the-journey-w…
Architecture Notes retweeted
I built more with AI agents in two months than the previous year. I used almost none of it. The real productivity gain isn't building—it's using.
mahdiyusuf.com/your-agent-is…
Architecture Notes retweeted
Why your coding agent keeps undoing your architecture
mahdiyusuf.com/why-your-codi…
Architecture Notes retweeted
Wrote something about the importance of the code generation and library use in the context of agents.
mahdiyusuf.com/agents-didnt-…
Architecture Notes retweeted
Building GenAI infra isn’t theory—it’s hard-won experience.
Learn from 60+ engineers who’ve scaled LLMs, vector stores & RAG pipelines at QCon SF 2025.
What worked. What failed. What to avoid.
Architecture Notes Weekly: Trio out now!
Why pgvector might not cut it in production
How bloom filters fail at scale
Why LLMs can’t replace hands-on learning
Plus: slashing cloud costs and Postgres as a lakehouse system
Join thousands of engineers! Don’t miss out 🚀
Building GenAI infra isn’t theory—it’s hard-won experience.
Learn from 60+ engineers who’ve scaled LLMs, vector stores & RAG pipelines at QCon SF 2025.
What worked. What failed. What to avoid.









