The next advantage in AI will come from systems that compound human expertise.
During his talk at @UChicago , @paulhsu explored how AI-native firms are evolving through specialized AI teammates, persistent context, memory, and continuous learning.
We're building toward this future at Decasonic.
2
94
Decasonic retweeted
great read. as consumer AI investors (@decasonic) we generally tend to view the landscape for investment opportunities as being defined by the ability of new products to present differentiated experiences.
those experiences are often defined by behaviors:
1/ existing and familiar: products tend to provide "automation," enabling (1) time or (2) cost-saving
2/ introducing new behaviors: leads to (1) enhanced defensibility and provides an opportunity to demonstrate (2) magic to users.
naturally the opportunities for "remixing," provides a reason for consumers to stay and increase product utilization (which is a natural byproduct of a network of users being onboarded). primary question is centered on the "time to magic," as the initial hook for users.
Great technology is built by resilient communities. And building together makes all of us more capable.
Our 6th Annual Frensgiving brought together Chicago's AI community to share ideas, strengthen connections, and explore how human expertise and AI can move us forward.
Grateful to everyone building alongside us, and to @startdancingai and Ferris & Co. for helping us make the event come to life.
Claude's new Whiteboards and Sheets put shared workspaces at the center of AI workflows.
The product question is whether teams repeatedly plan, review, and revise alongside the model.
If that behavior sticks, the interface becomes an important part of the value.
Starting to roll out today: Claude Whiteboards, Claude Sheets and Claude Data.
Think a plan through on a board, build a tracker in a sheet, and tell a story with your data.
102
Building together in the age of AI starts with bringing the right people into the room.
This Thursday, we're hosting our 6th Annual Frensgiving in Chicago alongside @startdancingai and Ferris & Co.
An evening of ideas, conversations, and connections with the community shaping what's next in AI.
See you on November 12.
luma.com/2b1wjs4c
Decasonic retweeted
there's really two main obstacles to agentic commerce today, which is resulting in an increasing amount of "walled gardens,"
1/ disruption of business models: advertising revenue. combination of the latest quarterly earnings from (1) google and (2) reddit point towards the disruption of the search models and increasingly reliance on leveraging AI to drive answers (minimize the time to intent). advertising is the lifeblood of commerce-related revenue.
2/ refunds: likely the biggest obstacle, and primarily is about ownership and liabilities. (1) there's a number of "layers," involved in agentic commerce: (1) users, (2) agent/interface of intent, (3) marketplace, (4) payment provider, (5) financial issuer. in the case of a misfire of an intent (as seen with the recent @Visa benchmarks) - who is the liable party responsible for delivering on a refund? this remains a very large question to be answered, despite the significant amount of fundamental work being made.
increasingly realizing that the "poppy standard," could truly be a solution to this. businesses provide a means for (1) agents to communicate with their business, and (2) this is likely the most standard approach - but it serves as a means of introducing A2A. this means you can encode a business's rules and limitations in an attempt to provide a solutions to the identified obstacles noted above.
Human expertise amplifies what AI can do.
@AnthropicAI found that expert users generate ~12 Claude Code actions per prompt, compared to ~5 for novices.
As AI agents take on more complex tasks, the ability to guide, evaluate, and refine their work becomes increasingly valuable.
Human expertise shapes how far AI can go.
2
2
128
Decasonic retweeted
part of the ease of software creation is the development of experiences at scale. often times what this leads to is the development of (1) one-time software, non-reusable and (2) standalone software.
assistant-first platforms similar to @Muse and @wabi could provide the ability for users to more efficiently utilize the software being created through:
1/ reduce tech debt from the need to maintain the many applications at scale. in theory, now it's one prompt to muse (or done proactively in the background)
2/ mitigate against hosting considerations (one click deploy. no need to figure out cloud considerations - similar to Replit and Lovable). of course this also includes security considerations
3/ enable remixing of software that could allow individuals to build and leverage on top of one-time software (system-level network vs individual standalone software). imo this is the crown jewel that leads to the ability to build network effects through multiplayer experiences.
As robots become more capable, coordination, specialization, and orchestration will become increasingly important.
@abdulalali highlights four developments worth watching, particularly as physical AI moves toward real-world deployment at scale.
there are parallels that we observe in software AI that will continue to emerge in robotics. some of those patterns will include:
1/ effective delegation of robot to robot collaboration. this primarily includes the ability to delegate tasks (sub-tasks) and goals
2/ human to robot collaboration. this includes both (1) delegation from humans and (2) wider collaboration on the ground.
3/ specialized models. post-training is accelerating (primarily due to @physical_int) and the release of API models @dreamlabs_ai. perhaps the key learning from software AI will carry over to rapid adoption of post-training specialized models across deployers.
4/ orchestrators responsible for delegating fleet management tasks. primarily draws on (1) but is truly a next "leg up," from the general fleet management software companies we see today.
where you land on this perspective is whether you believe there will be a (1) multi-hardware manufacturer and (2) multi-model future or a vertical integration from one core provider [a shining example of the latter being the figures of the world].
2
148
AI is building AI. Intelligence is compounding.
At Decasonic, AI agents co-authored 98% of our October code changes, guided by human expertise.
Our latest AI Ship Day recap explores what we're learning about recursive self-improvement.
2
2
280
Learnings from today’s @decasonic AI Ship Day: Trust in AI grows when systems connect observed evidence, explicit assumptions, and human judgment to verified scenarios.
We learned from today’s demos that making that process visible and capturing the expertise that improves our thinking about today and tomorrow.
A few takeaways:
1. Forecasts and scenarios work together. A forecast assesses what is likely to happen. Scenarios explore what could happen under specific conditions. Connecting observations to milestones, a base case, and alternative scenarios helps us understand which assumptions drive the assessment and what evidence would change it.
2. Physical AI simulations become more useful with richer sensor data. We are active in deal flow across this area. Collecting and synchronizing data can create more signal-rich simulation environments. That richness helps capture physical interactions, operating constraints, and failure conditions. The opportunity is to build environments where AI can test scenarios against a more complete representation of reality, with real-world validation guiding each cycle.
3. Human observation and verification build trust. Chain of thought is part of this discussion: a model-generated explanation can help people investigate an answer, but it does not necessarily reveal the model’s actual internal reasoning or establish correctness. Trust requires traceable evidence, explicit assumptions, and verification. Experts identify meaningful signals, challenge missing context, and test whether the stated rationale holds up against observed outcomes.
4. Expert feedback can become valuable post-training data. As we train our open-source models, expert demonstrations and corrections can support supervised fine-tuning. Comparisons between outputs can support preference training, while reliably scored outcomes can provide reinforcement learning feedback. Capturing the context and rationale makes that feedback more useful. Improvement still needs to be demonstrated on cases held out from training.
5. Human expertise improves what reinforcement learning rewards. Experts help define success, distinguish meaningful progress from superficial performance, and identify when a system exploits a scoring rule without achieving the intended result. Capturing their interventions, corrections, and outcome assessments can improve reward models and evaluation criteria. Preserving who contributed the judgment and under what conditions helps us assess where that feedback applies. This gives RL a stronger signal for learning behavior that holds up in practice.
AI augmentation becomes more valuable when human expertise is captured, tested, and applied across workflows, esp those related to the future decisions we need to make. The opportunity is a learning loop where verified outcomes strengthen shared context and inform the next training cycle.
Always shipping. Always learning.
Decasonic retweeted
increasingly, there's a clear-trend of companies emerging with the dedicated goal of selling real world deployment data (logs and traces - "data") to frontier labs. the business model is primarily predicated on determining the (1) value of benchmark improvements relative to the (2) cost of acquiring the data.
there's a very interesting question about the number of middle "layers," needed to provide the data back to the labs. there's a potential opportunity in the future where labs could offer "inference cashback," beyond the "promotional period," often seen on @OpenRouter and other inference marketplaces that could last and be durable. this cashback would be offered to businesses and users directly, with the labs guaranteeing the "anonymization," of user data.
will see if this emerges in the near-future, but there's clearly an opportunity for "capital markets," in the data layer to play a role.
AI is reshaping the economics of game development, enabling smaller studios to build and iterate faster with significantly fewer resources.
Our portfolio company @DPS_Studios is putting this into practice with Valesong, launching October 29.
John Smedley (@j_smedley) spoke with Dean Takahashi at @GamesBeat about the studio's shift toward AI-native development and what it means for the future of gaming.
gamesbeat.com/distinct-possi…
the frontier is
multiplayer networked,
multi-interface across physical and digital,
multinational in ambition,
with AI and blockchain integrating intelligence and ownership.
stay ambitious
Physical AI needs more transparent benchmarks as robots move closer to real-world deployment.
Making all 120 Dream-1 robot runs publicly available is an important step toward understanding where these systems perform, where they fail, and how reliably they can operate.
We believe rigorous, reproducible evaluations will be critical to building trust and accelerating Physical AI adoption.
Replying to @robocurve
We release all 120 Dream-1 robot runs, with full transcripts and videos: robocurve.org/all-runs/?lead…
2
166
Decasonic retweeted
incredibly excited for the adoption trajectory of @robocurve.
the physical AI benchmarks and evals market will note a similar trajectory to the software AI benchmarks (from generalized tasks to economic tasks). as this continues to accelerate, and as deployment continues to be a crucial focus of my companies - a standard benchmark is needed to assess the performance of models in the real world.
Check out the full 120 eval rollouts done by @robocurve . Full transparency is the only way to move the industry forward.
AI agents need infrastructure to discover services, execute tasks, and transact independently. As more economic activity becomes agent-driven, secure payments and trusted access will be critical to adoption.
Great to see @orthogonal_sh alongside the growing ecosystem building on @privy_io.
Proud to back @orthogonal_sh in advancing the infrastructure for agentic commerce and making autonomous transactions possible at scale.
Today is probably a good time to check in on things with @privy_io, which became part of @stripe just over a year ago.
Our acquisition was predicated on three factors: stablecoin growth, self-custody as a key enabler of broader global coverage, and rapidly increasing demand for programmable financial infrastructure. As we said at the time, "money has to reside somewhere, and Privy builds the world's best programmable vaults."
Since then, the Privy team has executed exceptionally well: assets in Privy wallets have grown ~25x, monthly active developers are up ~3x, revenue is up ~9x, and Privy now facilitates nearly $25B of monthly transaction volume. @tryramp, @Deel, @Klarna, @felixpago, @HyperliquidX, and thousands of other businesses are building embedded banking functionality on Privy.
At Stripe, we're deeply integrating Privy into Treasury, @Link, and Connect. In each case, it's accelerating our global expansion, and enabling us to build better functionality faster. (Coming soon: @Link balances.)
We're also seeing that Privy's wallet infrastructure is the ideal substrate for agents to transact and earn -- live today with @merit_systems, @orthogonal_sh, @Nevermined_ai, @prlnet, and more. Developers prefer programmable financial infrastructure, but agents need it. We expect that the agentic economy will continue to grow extremely quickly.
The Privy team has maintained remarkable product velocity, with major launches this year including deposit addresses, earning on Privy balances, digital asset accounts, custodial wallets, and Stripe's own stablecoin cards.
It's also been neat to see how Privy has been accelerated by Stripe, with better onramps, @stablecoin payment rails and licenses, Stripe's fiat capabilities and issuing stack, and integration with @Tempo.
Together, we're building one API for holding stablecoin balances and earning yield, moving money in and out of fiat, and orchestrating spend with cards. We still have a lot of work to do, not least in integrating everything across Privy, Bridge, and the rest of Stripe -- there's been so much feature and functionality expansion that we haven't yet had much time to reconcile everything. Overall, though, I'm very happy with how things are going, and excited by where these capabilities are headed.
If you're interested in working on the future of stablecoins, agentic payments, and crypto's deployment in the real world, you should get in touch with @sternhenri (now leading all of @StripeCrypto) or others on the team.
29.1 million web requests in just 12 days. Meta Muse is offering an early glimpse into how consumer AI assistants could change the way people interact with the internet.
8.9% of sampled requests touched transactional or account-related endpoints, including forms, logins, and payments.
At Decasonic, we believe the next phase of consumer AI adoption will be driven by assistants that can execute within everyday workflows. Persistent context, trusted access, and the ability to act on behalf of users will become increasingly important as AI moves deeper into consumer behavior.
2
126



















