Introducing PorTAL: Portable Task Adapters for LLMs. A novel recipe to cheaply port fine-tuning between models. It matches per task LoRA accuracy at half the cost, lowering the switching overhead of adapting tasks across LLMs.
At Ramp, every new model release used to mean retraining our fine-tunes from scratch. PorTAL learns the task once, then efficiently refits it onto any new base model, even across model families.
VVV and Capital
Measured by revenue, Venice has become the largest company at the intersection of AI and cryptoeconomics.
Today, we announced Venice’s first round of outside capital, a $65m Series A led by @dragonfly_xyz, valuing Venice’s equity at $1 billion.
Since we are an atypical company, this thread describes how this equity raise relates to the most valuable asset we have on our balance sheet, our capital token base:0xacfe6019ed1a7dc6f7b508c02d1b04ec88cc21bf
You don't understand how BIG this is.
Until now, agents could search the web but X was basically a wall. Real-time posts, trending topics, what people are actually saying right now, none of that was easy to pull into an agent workflow without a complicated API setup.
X just shipped a hosted MCP that changes that. Connect Grok, Cursor, or any MCP-compatible tool to the X API with no setup at all.
What this actually means if you run agents:
→ Your content agent can watch what's trending in your niche and surface it before you open your feed
→ Your research agent can pull real-time reactions to a product launch, not just news articles
→ Your morning brief now includes live X signal, not just web search
I run a content operation across 5 businesses. The bottleneck was always "my agent doesn't know what's happening right now." That just got solved.
If you're running agents and not wiring this in this week, you're leaving real-time context on the table.
Merry Christmas, here's some genuine free alpha that you can use
Funding Rate vs Realized Drift
automation not required
most of y'all think of the funding rate like RSI; negative = oversold, positive = overbought
but you're wrong, it's way more specific
Jeff Bezos on his favorite interview question
“When I interview people, I ask them to give me an example of something they've invented. And I always point out, it doesn't have to be something that you actually took to the patent office. It could be a metric that you invented and followed carefully. It could be a business process that you invented. You want to select people who like to invent their way out of boxes.”
A lot of people will immediately jump to what Jeff calls an “either or” solution (e.g. “We can do A or B.”). But as Jeff explains:
“The right question is how can we do A and B? What invention do we need to be able to do both?”
my ideal AI interface is a single never-ending chat thread.
i don't want to think about the concepts of sessions, context windows, worktrees, mcp servers or really anything else.
The harness should automate everything transparently.
Karpathy just wrote the manual for Claude + Obsidian as a real second brain.
Most vaults die the same way. A year of saved articles and highlights. None of it linked. The graph rots while it still looks impressive.
So he moved the upkeep to the model. You curate sources and ask questions. Claude files, links, and reconciles. You keep judgment. It keeps the books.
raw belongs to you and never gets edited. wiki belongs to Claude. It isn't RAG. Your sources compile once into linked pages and compound from there.
9 rules. Start with 10 sources, not 10,000.
Most people hoard notes. This turns them into a brain that maintains itself.
Meet Hermes Agora
A 3D command center for Hermes Agents
Built 100% by Hermes Agents connected to Unreal Engine
Your AI agents now have a place to live, think move and collaborate.
Coming soon!
Just put my systematic trading handbook live on my website.
24 chapters, start to finish: what an edge actually is, building in MT5, backtesting without fooling yourself, the overfitting trap, position sizing and the maths of ruin, then going live and managing capacity.
I also dropped a free resources page: the MQL5 order-splitting code I run live in XAQP, which cuts market impact and lifts a strategy's capacity ceiling.
I wrote it all as the thing I wish someone had handed me when I switched from discretionary to systematic.
Link in bio.
If you want to read more of my work, please consider subscribing to my Substack for weekly publications.
in @ycombinator they have a playbook on how to get customers ASAP for your startup.
if you follow this, you’ll brute force your way to 100 customers, almost no matter what your product is.
Here it is:
1/ launch-max.
product hunt, hackerNews, devhunt, betalist, peerlist, indie hackers, etc. YC tells you to launch 3 times MINIMUM
2/ pull your competitor’s strongest backlinks and get yourself listed in the same places.
whatever article they have listed, you make a better version and ask the site to replace it (or supplement) with yours.
3/ WARM OUTBOUND.
Everyone knows about building in public. but you still need to capitalize on the 99% of leads who see your content but don’t come inbound
scrape everyone who likes your posts on Linkedin each week, check if they fit your customer profile, and message them.
you set this up to fire automatically with @origamichat (i dropped a prompt in the comments)
4/ find 20 to 30 ugc creators on tiktok / instagram in your niche. ask them to create content about your product, ideally from a fresh account.
pay them a fixed fee ($15–$30 per video) plus performance incentives ($1k for 1 million views, etc).
you can use @sideshift_app (best creators imo) and line up 20+ of these creators in 1 day
5/ when building in public, a video is 10x better than an image/text - spam use cases of ur product on X/Linkedin
6/ figure out where your customers actually spend time.
which slack/discord groups are they in? what newsletters do they open? which podcasts and accounts do they follow? pay those people for shoutouts
7/ there's a fresh trend on x basically every week. jump on the relevant ones and fold your product in (like i’m doing right now).
To find trends i just use Origami & search “Lead Gen/GTM posts that are viral on X” to find the best posts every week in my niche
Then, I will reply to those, quote tweet them, and use the formats that work myself
(that’s the secret to why my account has high engagement BTW - you can do this too)
---------
if you are doing all this every single week and DO NOT GIVE UP (launching, posting demos, contacting new customers)
I guarantee you will hit your customer goals. Then the game becomes retention.
will be posting 2-3 more growth hacks every single week
Hermes Agent now supports asyncronous subagents!
The existing delegate tool, which your agent uses to spawn subagents to fan out and do work, no longer blocks your chat!
To access now, `hermes update`, and enjoy!
In partnership with @stripe, Hermes Agent now supports a full suite of Stripe skills.
Your agent can buy things, pay per-call APIs, and provision its own SaaS, with configurable safety limits on every action.
Karpathy said something you'll regret ignoring:
"Remove yourself as the bottleneck. Maximize your leverage. Put in very few tokens, and a huge amount of stuff happens on your behalf."
Loop engineering is the exact thing that does that.
In a hand-run session, the operator handles two things:
- deciding what the agent runs next
- and checking its output before the next step
Both are manual, and both decide how far the agent gets on its own without the operator.
Loop engineering moves both steps into the system.
A core operating structure surrounds the loop, and the diagram below depicts it.
- A schedule decides what to run
- Loop is the maker that produces the work
- A separate checker agent grades the output
- A file on disk holds the state they both read.
The loop runs until either done, max iterations, or an exhausted budget.
Here are some practical engineering considerations:
1) A model grading its own output justifies what it already did instead of catching where it failed.
That's why a separate checker's findings return to the maker as the next instruction. And the cycle repeats until the checker finds nothing left to fix.
2) A loop with no stop condition burns tokens, and the cost climbs fast once sub-agents and long runs add up.
That's why the exit must be set before the loop runs, not while it is running.
A simple exit could be:
↳ fix only the major issues, run one final pass, and stop after two loops, with "all tests pass and lint clean" as the rule that ends it.
3) State has to live on disk, not in context.
The model forgets everything between runs, so an MD file or a knowledge graph holds what is done and what is still open.
Each run reads it and writes back to it, which lets a loop pick up again after days.
4) The lower the verification bar, the safer the loop.
Boring, repetitive checks like a stale version string or a missing test are trivial to verify, so a loop runs them with little risk while the operator is away.
Judgment-heavy work is loopable too, but only as far as the checker can confirm the result.
Let's look at how an unattended loop fails in two ways.
1) It reports done when nothing is actually verified.
The separate checker exists to prevent it, but it merges code faster than anyone reads it, so over weeks, the team stops understanding its own codebase while every check stays green.
Green tests say the code passed the tests, not that anyone knows what shipped. Someone still has to read what the loop merges.
2) The checker keeps a running loop honest, but it only catches failures inside a run.
The harness around the loop, like the prompts, tools, and checks wrapped around the model, still drifts and breaks in production as models change.
That repair loop is usually run by hand based on observability traces.
My co-founder wrote a detailed walkthrough (with code) on making that harness repair itself, where a failing trace gets diagnosed, the fix is verified against the exact input that failed, and the failure is locked as a regression test so it cannot recur.
Read it below.
My favorite @elonmusk quote that I often send friends:
Do not fear losing. “You will lose,” Musk says. “It will hurt the first fifty times. When you get used to losing, you will play each game with less emotion.” You will be more fearless, take more risks.
it's not done if it's not implemented
it's not done if the implementation is ugly
it's not done if it's not documented
it's not done if users can't discover it
it's not done if you can't market it