krisoshea.eth (@krisos) header
krisoshea.eth (@krisos) profile picture

I like blocks | @ethereum Protocol Fellow | Prev: @OlenaProtocol @SumeroApp

Decentralized
Joined May 2009
Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)Photo by krisoshea.eth (@krisos)
krisoshea.eth retweeted
I predict in 12 months there will already be a large gap between system programmer's who learned FV and Lean and start using it and those who didn't. The things you'll be able to unleash your agents to do with proofs that you can't without will be enormous
36
56
7
1,448
78,705
krisoshea.eth retweeted
The Lean FRO and Mathlib Initiative have released a joint statement on the Lean verification of the Riemann Hypothesis. On November 20, Levent Alpöge released a proof of the Riemann Hypothesis as a Lean library of 17 million lines of code. Over the past day, the Lean and Mathlib communities worked together to verify it. The statement certifies the proof as correct and describes how it meets all necessary verification criteria. Congratulations to Levent on this achievement, and thank you to everyone in the community who contributed to the verification. Full statement: leanprover-community.github.…
550
5,320
2,084
28,587
3,414,125
krisoshea.eth retweeted
For the 10,000th time proof != spec The magic of Lean4 and other tools like it is that while e.g. the Riemann Hypothesis proof is ~17M lines of Lean4 you only need to read the spec, ~16 lines of Lean4 + trust the kernel to know it's correct
Replying to @banteg
What if there’s a mistake in the spec?
22
18
2
680
53,901
krisoshea.eth retweeted
BREAKING: Robert Pattinson will play Claude in an upcoming biopic on the Riemann Hypothesis.
117
3,134
171
48,893
747,596
krisoshea.eth retweeted
90% of backend engineering in 2026 is mastering these 10 concepts: 1) Latency budgets Know p50/p95/p99 and where time goes (DB, network, serialization). Most outages start as slowdowns. 2) Data modeling + transactions Pick the right constraints and isolation level. Many bugs are really duplicate writes, lost updates, or missing unique indexes. 3) Concurrency + backpressure Bound queues, time out work, shed load. Unlimited goroutines/threads and unbounded retries turn a blip into a pileup. 4) Idempotency + retries Assume clients, jobs, and webhooks will retry. Use idempotency keys, dedupe tables, and exactly-once illusions sparingly. 5) Caching tradeoffs Cache invalidation is hard, but stale reads are a product decision. Decide TTLs, cache stampede control, and what must be strongly consistent. 6) Deploy safety Health checks, rollbacks, feature flags, and schema migrations that support old+new code. Most bad deploys are contract breaks, not logic bugs. 7) Observability with intent Traces for shape, metrics for paging, logs for proof. Correlation IDs and high-cardinality labels matter more than dashboard count. 8) Debugging in prod Know how to bisect a regression, sample traffic, reproduce with real inputs, and inspect live state (profilers, heap dumps, slow query logs). 9) Security defaults Least privilege, secret rotation, input validation, and dependency hygiene. Most incidents are authz mistakes and leaked credentials, not crypto. 10) Failure mode thinking Design for partial outages: timeouts, circuit breakers, bulkheads, and degradation paths. Treat every dependency as unreliable under load
6
15
152
6,982
krisoshea.eth retweeted
With ePBS (enshrined proposer-builder separation, EIP-7732) live on Sepolia, we measured the builder market’s first month: 240,632 slots from the Gloas fork to 8 Nov. Our first read on who builds, how concentrated the market is, and the strategies builders use: - External builders now build ~80% of slots, up from 51% in week one. Counted by operator, the market is tighter: 3.5 effective builders by index vs 2.5 by operator (HHI). The largest operator runs five indices. - Why run several indices? Gossip forwards one bid per builder index per slot and parent. A builder with five indices reach our node from ~140 ms before to ~920 ms after the slot start. Each later bid can carry a later, fuller block. - 5% of builder blocks were won with trusted bids. These pay inside the execution payload, never appear on gossip, and the protocol doesn’t enforce their payment. Three of them left proposers unpaid. Check the full article here and share your thoughts with us: migalabs.io/blog/epbs-builde…
1
3
3
15
523
krisoshea.eth retweeted
Replying to @miga_labs
@miga_labs published builder data of the first Gloas month on Sepolia. I worked with @krisos on the consensus side of builder.nflaig.eth (@lodestar_eth builder). Running on @nicoflaig’s Zig building engine, it became the largest operator, building 44.5% of builder blocks.
With ePBS (enshrined proposer-builder separation, EIP-7732) live on Sepolia, we measured the builder market’s first month: 240,632 slots from the Gloas fork to 8 Nov. Our first read on who builds, how concentrated the market is, and the strategies builders use: - External builders now build ~80% of slots, up from 51% in week one. Counted by operator, the market is tighter: 3.5 effective builders by index vs 2.5 by operator (HHI). The largest operator runs five indices. - Why run several indices? Gossip forwards one bid per builder index per slot and parent. A builder with five indices reach our node from ~140 ms before to ~920 ms after the slot start. Each later bid can carry a later, fuller block. - 5% of builder blocks were won with trusted bids. These pay inside the execution payload, never appear on gossip, and the protocol doesn’t enforce their payment. Three of them left proposers unpaid. Check the full article here and share your thoughts with us: migalabs.io/blog/epbs-builde…
2
5
111
krisoshea.eth retweeted
The computer science major is going through an identity crisis. Claude Code can finish any programming assignment with a single prompt so what's the value of teaching students how to write a function in Python? Here’s the point: we still need architects, not button-pushers. The next decade will belong to people who understand theory and how to break complex systems into smaller components even without Wifi. Imagine a degree that drills algorithmic thinking. Weekly closed-book exams. CLRS becomes the most important book in the major. And software development? You’ll exercise that muscle only at internships and real jobs. The computer science major will look more and more like a math degree.
74
123
25
1,095
69,901
krisoshea.eth retweeted
Swarm riders and Swarm-native companies AI-use even among highly AI-native people is power-law distributed. There seems to be a new class of people — call them “swarm riders” — who are capable of consistently commanding large swarms of agents to productive ends. This is a new kind of managerial skill, analogous to the skill it takes to manage large organizations of people, but wholly different in nature and of course more technical. Based on the swarm riders we know (including many at @Tempo and @Paradigm), these people are capable of productive token spend that is 1000x-10000x higher than median software engineers, and 100x higher than top 5% Claude Code users. The art of swarm riding is different from software engineering and more akin to machine learning, where practitioners grapple with black box systems: defining reward functions and evals, transmuting non-verifiable problems into verifiable problems, running experiments, etc. Swarm-native companies are a new class of company founded by and populated by swarm riders. These companies are organized around swarms in a way that conventional companies will struggle to replicate, as the cost of internal human coordination is sand in the gears of an effective swarm. Effective swarm-native companies will define the right reward functions and evals, remove functional and bureaucratic barriers, and then let the swarms cook. We’ve gotten to see a few swarm-native companies up close — reshaping financial rails @Tempo, building a new gaming platform @Spawn, cyber swarms at Cathedral Systems — and it’s clear these teams wield a significant advantage over pre-AI teams. Some questions I’ve been wondering about: 1) How fast will swarm riding diffuse through the economy? The implications for token spend could be significant. 2) Can swarms be effectively adopted by existing companies or will it generally track the formation of new swarm-native companies? 3) What becomes scarce as swarm capabilities themselves commoditize (through frontier models becoming better at swarming, swarm capabilities built into harnesses, etc)? Cultures and teams that can define reward functions well and otherwise get out of the way?
145
83
116
1,184
285,136
krisoshea.eth retweeted
"Søren Kierkegaard wrote that life can only be understood backward, but it must be lived forward. We’ve encountered a similar bind when designing memory for AI agents." Absolute banger opener.
Memory is what allows Claude to learn from experience. In a new Engineering Blog post, we share how “dreaming” helps Claude organize what it’s learned, correct its mistakes, and find what it needs more easily the next time. anthropic.com/research/compo…
21
127
6
2,383
83,865
krisoshea.eth retweeted
I've changed how I work over the last few weeks. I now have one lead agent for each area of focus, running all the time, and a chief of staff agent that oversees them. Each lead manages its own sub-agents that write code, test it, chase reviews, and coordinate with team members. Everything I need to decide or review comes into one prioritized queue, a simple local web app. Each product review includes screenshots and a video demo of the feature. I clear the queue in batches, mostly by voice and keyboard shortcuts. The leads also bring back ideas I hadn't considered. Working from one queue, instead of switching between dozens of chats, has cut a lot of mental overhead. The system keeps learning my preferences and distilling my judgment into a "Brian brain," so its recommendations and first drafts get better before they reach me. The goal is fewer corrections per change. Every change still goes through the same human review and security checks. It is eliminating tasks, not human judgment. What an exciting time to push the frontier of productivity, especially as we build the economy for agents (agentic finance). We should all keep pushing in this direction.
871
479
437
10,283
1,668,542
krisoshea.eth retweeted
hot take: you don't need a software factory, and i don't love the term. it's just marketing all you need is good software engineering and investing your tokens in the right places. the more deterministic the better (and token efficient)
all the companies selling you a software factory produce a shockingly tiny amount of software
439
587
183
8,164
521,043
krisoshea.eth retweeted
People keep asking how companies like Ramp, Linear, and Anthropic ship so much software with agents. There is no new SDLC or secret set of tools. The tools are increasingly available to everyone. You can build your own, use Linear, Claude, or whatever works. The bigger difference is ambition. These companies encourage their teams to build more, experiment more, and let agents do more of the work. That's the main different between "AI native" or not. You try to use the technology to the max. But don't get there by adopting a tool. You get there by changing what you believe your teams are capable of. So in the very simplest way: - let agents build - make it easier and faster to contribute - increase the ambition of what your teams attempt. In practice, that means things like: - Standardize around cloud agents. Most of our code (5M lines last quater) is now written in Linear with our Linear cloud coding agent. Cloud sandboxes make standardizing the system and environment easier for everyone and enables faster contributions - Linear agent (with code access and cloud agent) operates in Slack too. If there is a discussion or someone notices a problem just say "@linear fix this" (decrease the time from something noticed to fixed) - As bugs, errors etc arrive in Linear, send them to Triage, and have Loop to auto-investigate (connect to tools like Sentry, Datadog), and write a fix - Let designers, PMs, non-technical people contribute code where useful. If they notice a problem, tell them they can submit a fix. If they want to try something, ask them to build a prototype. - Set up code review agent and system to make sure PRs are safe to merge. - Run security as an automation, have agents constantly try to find security issues in your codebase, as they find, send it to a coding agent - Instead of working on issue level, increase ambition, make a whole project, ask the agent to ask questions, investigate code base, create issues and have the agent work on the whole project at once. Set up goals like "give me proof that what you build works". All our PRs now come with screenshots and videos and preview build you can - Bring customer feedback to the system like Linear, and make it available for your team. At Linear every project now starts with 10-500 customer notes that influence the direction of the work, and which customers we should ask to test the solution first. - Set up Loops/automations to constantly update reference documents for agents, like product area/code owners so work can be routed to right humans Then what is left is how do you set direction for your teams (initiatives, projects they should work on) and how do you review results (better code review, Linear does it with Guided Reviews), how you maintain quality/cohesion which is not a new problem either. Basically most of the problems you always had remain, but the execution bit is now more and more solved. At @linear we can help with automation, cloud agents, and code review, customer feedback and planning bits. But we can't change your mindset or how your company operates. If your company is still spending energy debating jira workflow steps, you're probably solving the wrong problems.
145
59
46
1,250
100,486
Anthropic @AnthropicAI ❤️ zkSTARKs: In collaboration with @StarkWareLtd co-founder Prof. Alessandro Chiesa and other brilliant researchers, Anthropic just open sourced a zk STARK system for proving the integrity of LLM inferencing, bringing us closer to verifiable AI. As we said long ago (since 2018), zkSTARKs are the end-game: highest scale, best security (post-quantum). Now, also AI agile and supporting AI integrity. anthropic.com/research/prova…
58
124
32
814
56,046
krisoshea.eth retweeted
well well well, things I didn’t expect: • anthropic shipping zk proofs for ai inference • with that ofc the proof system is similar to what ethereum rollups use (so surprising to see which rollups got mentioned) cc @gluk64 • part of the jump on speed is that they used h100, the earlier proof of the same model was done in 14 hrs for 1k tokens, compared to 9sec today anthropic.com/research/prova…
24
44
23
414
59,648
idk how they did it but it's literally black magic 9 MINUTES to compile on solc 2 SECONDS on solar and bytecode is ~40% smaller, gas is ~15% cheaper holy fucking shit
SOLAR: 100x FASTER SOLIDITY Solar is the future of Solidity and the EVM. It's >100x faster than Solc. It has faster bytecode than Solady with no assembly. Its bytecode is faster to interpret by the EVM. Solar fixes Stack Too Deep! foundryup forge --use solar Blog below.
14
18
5
282
19,354
SOLAR: 100x FASTER SOLIDITY Solar is the future of Solidity and the EVM. It's >100x faster than Solc. It has faster bytecode than Solady with no assembly. Its bytecode is faster to interpret by the EVM. Solar fixes Stack Too Deep! foundryup forge --use solar Blog below.
107
175
74
1,196
139,435
krisoshea.eth retweeted
"Your lack of urgency is wasting your potential."
You have to understand. The longer you wait to build skills, the harder life pushes you to settle for something beneath your potential capabilities. Education is a race against premature convergence.
31
1,529
51
11,208
529,998
krisoshea.eth retweeted
You have to understand. The longer you wait to build skills, the harder life pushes you to settle for something beneath your potential capabilities. Education is a race against premature convergence.
32
820
27
6,288
383,937