Founders Stage Podcast @outpostcast_ | #10 @NearLegion | Contributing @SuperteamTH
ITERA
Joined August 2025
- Tweets15.9K
- Following982
- Followers934
- Likes47.4K
simonbebek retweeted
We are at Security Armageddon now.
A large part of internet is affect in some way today, lots of older software vulnerabilities are coming out and through those more software is getting exploited.
PHP systems like Wordpress power a big part of content websites. Malicious TLS certificates especially affect basic services and older codebases. Plus node.js, plus LiteLLM, etc etc.
Uncensored models are becoming a key to enable security defenses to keep up. Formal verification allows to improve on how things are verified but still requires smart models to translate between systems and make sure formalizations are correct.
We also really need to transition to models that involve cryptography where right now API keys and access tokens are used. Hosting platforms that can secure the software even from platform itself.
Meanwhile, update, patch, rotate.
Your LLMs are weebs by delfauts, prove me wrong.
Researchers proved every LLM is secretly obsessed with Japan.
And they finally figured out why.
They published a paper investigating hidden cultural biases in large language models.
they tested 31,000 questions across 24 different languages on models like gemini, chatgpt, claude, and llama.
they stripped out all geographic hints and just asked the models to pick a place and answer.
the finding was universal.. across a wide range of models and languages, there is a pervasive "obsession" with japanese culture.
even when you prompt them in a language with zero ties to japan, they default to japanese norms, values, and aesthetics at a disproportionately high rate compared to its actual global demographic weight.
the craziest part? they tracked down exactly when the ai develops this obsession.. and it is not from reading the raw internet.
during pre-training, the models actually have a fairly balanced, global view of the world. the obsession gets injected during the safety phase.
when tech companies fine-tune these models to be "helpful, harmless, and polite," the ai starts mapping those corporate safety guidelines directly onto japanese cultural concepts. in the pursuit of creating the perfect, inoffensive, compliant assistant, they warped how the models view the world.
the quantization process (compressing models to run faster locally) actually amplifies this determinism. the smaller and faster your local model is, the more stereotypically it behaves, causing japanese cultural tropes to spike hard.
we spend all this time worrying about existential ai safety, but the real alignment problem is that our most advanced reasoning engines are hyper-fixated on japanese internet culture.
every model upgrade is reinforcing an invisible regional bias that dictates how they generate stories, facts, and code. the cultural commonsense of the internet is not universal..
it’s just localized to whatever the heaviest reddit and anime wiki users were obsessed with.
simonbebek retweeted
A lot of people aren't underqualified
They're just unknown to the people making decisions
The founder hiring for a role doesn't know your name
The client with a budget hasn't seen your work
The creator looking for a partner has never spoken to you
And someone else gets the opportunity
Not always because they're more talented
Sometimes they simply made the first move three weeks earlier
You don't need to dm every big account you see
Pick people whose work you understand, find a genuine reason to reach out, and bring something useful to the conversation
A thoughtful msg won't guarantee a yes
But staying invisible guarantees they won't consider you for an opportunity they don't know you want
What's the best opportunity you've ever got from reaching out first?
simonbebek retweeted
We answered crypto’s most debated questions!
00:00 Intro
01:05 Memecoin trading: skill or luck?
03:17 Brand building: skill or luck?
04:58 Talent vs hard work
05:59 Surviving the periods with no progress
07:51 Chase the narrative or stick to your guns
09:17 Is the world becoming more PvP?
13:25 What leverage actually means
17:44 The rat race, redefined
20:15 What actually drives us
22:46 Traders are the new celebrities
25:50 The dangers of FOMO
28:27 The one skill that made us successful
31:16 Real relationships in a transactional industry
32:45 Will AI replace human content?
33:25 Do brands still need founders with personal brands?
36:29 Rebuilding from zero in 12 months
39:25 Breaking through on a hot narrative
40:47 Your friends aren't obliged to support you
43:25 The solo years
46:12 Belief, mentors, and the right people
48:58 What's next?
simonbebek retweeted
🗳️ i voted FOR reducing near:native emission to 1.6%
addressing one common concern that says: “too many people will unstake and security/yield will collapse.”
this is a multi-disciplinary proposal, with many implications to @NEARProtocol and its ecosystem (i'll try to address some diff implications in the next few days)
so, on this "unstake" concern (which is a fair one) ↓
NEAR has a (clever) self-balancing mechanism:
lower emission reduces the reward pool, but any unstaking raises the effective APY for those who stay (same rewards shared among fewer stakers).
game theory: that higher yield pulls capital back in.
worst-case basis @ the 1.6% floor (90% to stakers = 1.44% of total supply to the staking set):
• baseline (42.71% staked) → ~3.37% gross APY
• 10% of stakers unstake (ratio → 38.4%) → ~3.75%
• 20% unstake (→ 34.2%) → ~4.21%
• 30% unstake (→ 29.9%) → ~4.82%
• 40% unstake (→ 25.6%) → ~5.62%
to fully restore the current ~5.27% APY you’d need ~36% of current stakers to exit (staking ratio down to ~27%). even then the remaining stakers earn more, which incentivizes re-staking. the system equilibrates.
this looking solely to APY measured in NEAR, not USD. if NEAR price increases against the dollar, the effective dollar-based APY compounds into a much larger ROI.
this is the same dynamic that already showed up recently: staking ratio fell while price rose and gross yield increased.
lower dilution + higher real yield for committed capital is net positive for long-term holders and for the network’s store-of-value case (talked about that in the quoted post).
but it's important to say that the benefits goes beyond speculative APY. validators, stakers, and delegators also have other intrinsic motives to stake beyond pure reward farm (keeping the network secure, for example, which is the infra for their businesses, projects, holdings, and settlement network -- to mention a few)
i’ll keep supporting the second step (validator upgrade) once the vote passes with thecoding.pool.near (currently with 420K+ NEAR staked and a self-sustainable node thanks to community delegation that will keep operating @ 1.6% tail emission
i just voted FOR reducing near:native emission to 1.6%
this is a multi-disciplinary proposal, with many implications to @NEARProtocol and its ecosystem (i'll try to address some diff implications in the next few days)
from an economics perspective: it opens doors for positioning NEAR as a strong store of value and alternative programmable money against the US dollar (and even against gold itself)
right now (October 2026):
• NEAR inflation is 2.5%
• US CPI is 3.7% (core is 2.6%)
• gold annualized supply growth is 1.5-2.0%
the target for the next 24 months:
• NEAR inflation to 1.6%
• USD inflation target by the FED is 2%
• similar expected growth rate for gold (1.5-2%)
of course, there are some nuances worth mentioning here.
according to the Austrian Economics School, inflation measures the monetary supply of a currency, not the consumer price index (CPI) increase that is a natural effect of an inflating currency (that gets diluted and loses its purchasing power over time)
but we can't really know how much the FED is printing, so the CPI is what is publicly available to measure the dollar inflation (highly prone to error, imo)
loss of purchasing power can also be affected by other factors, as, for example, with supply & demand elements related to the products and not he currency itself
the same thing is true about NEAR increase/decrease of purchasing power (or its price against the USD). supply dynamics (NEAR emission/inflation) is one factor of this equation.
the demand for NEAR (utility, speculation, etc) directly influences on whether it may or may not become a strong store of value. microeconomics (revenue, fees, burn, etc) + macroeconomics also have a high influence on that
however, supply dynamics (emission) IS STILL an important factor and it's a great step into making NEAR sound programmable money that can also act as a store of value people want to acquire, hold, stake, and use
i vote for the part that is on our control
we can't control demand part (we can create incentives for it, but not control). we can control the supply part via this proposal
i'll also support the second part of this proposal, that is upgrading my node (thecoding.pool.near) to the protocol update that will be built if this vote passes on @NEARGovernance -- as an approval on HoS will still require 80% of validators to upgrade, being the second voting round
FUN FACT: gold itself has an annual supply growth rate between 1.5-2%
Break- Point.
Dear UK Founders,
Your friends think you’re living the dream.
Your family still asks when you’ll get a proper job. You constantly think no-one understands what this vision actually cost you. Some weeks you’re not even sure yourself.
You’ve redone your deck 100 times. You’ve had VC calls where the investor said they loved your product then they ghost you.
Meanwhile, half the people you were building with have moved to SF because “that’s where the money and momentum is”.
But you stayed.
Some days you wonder if that was the mistake.
It wasn’t.
For 3 days in November, the momentum comes to you.
People from 100+ countries will gather, connect, close deals and learn.
There will never be talent this dense in London again.
This opportunity moves to a new city every year, which means you’d normally need flights, visas and a lucky intro to the one person who can change your startup's trajectory.
But for one weekend only, it'll just be a Tube ride away.
The meeting you’ve been chasing for months will just happen. The investor who went quiet on you will sit in the same audience as you.
For once, you’ll be surrounded by thousands of people who understand exactly what you’re going through, because they’re going through it too.
You’ve bet everything on what you’re building.
Why not spend 3 days in the room where people will actually see its potential, and maybe even back you?
You know where my position at in this case.
Feel free to go against me.
near:native inflation cut is now up for a vote 🗳️
HSP-029 is live on House of Stake: max issuance 2.5% to 1.6% over 24 months.
But a vote isn't a cut. It still needs a nearcore release + validators to upgrade.
Will it be live on mainnet by April 6? Traders say 57% 👇
noamarket.xyz/markets/8-will…
simonbebek retweeted
Objectively, ZachXBT is the main protagonist of CT.
Don't care if you think he's a psyop or some sort of government agency, he's a crime hunter.
Happy to have outpostcast into the nexus node @nexusolapp.
Streaming and reach like-minded founders is the easy part, what actually matters is "asking the hard questions" which often time forgotten.
@dulfariz_ @SuperteamTH
Welcome to Nexusol, @outpostcast_ 🎙️
A weekly founder podcast supporting the @SuperteamTH community, streamed live as Outpost Talk. One Solana team per episode, candid questions, no shill bookings.
Recent guests: @AccriumFi, @watchverta and @chordexpress, all on their page and NOW on 🔽
nexusol.app/app/ecosystem/pr…
