Stefano Fantini Delmanto
Founder of Oasis, The AI-Native Organization built by Ex-Palantir FDEs
Company
Stefano Fantini Delmanto is listed as a founder of Oasis. The AI-Native Organization built by Ex-Palantir FDEs Speedrun 006 profile: 5 employees; located in New York, New York; tags: Media / Entertainment / Creator Economy, Infra, Enterprise SaaS.
Founder traction evidence
Public posts and activity attributed directly to this founder.
- X
Oasis virtual-office launch
Your next coworker won’t be human. Introducing Oasis: The first virtual office where humans and agents work together. We’re building 1,000 agent teams for free. Comment what you need and we’ll build you an army of agents.
- X
Join Us!
Join Us!
- X
One of the hardest UX problems in AI is making agents feel real without making them feel phony.
One of the hardest UX problems in AI is making agents feel real without making them feel phony. Most interfaces lean on avatars, characters, and game mechanics. It ends up feeling gimmicky instead of believable. @jaezun_ has done an amazing job navigating that balance.
- X
Working on truly collaborative artifacts with @jaezun_ was so much fun
Working on truly collaborative artifacts with @jaezun_ was so much fun
- X
AI spend should be outcome-based.
AI spend should be outcome-based. Your AI bill tells you how many tokens you used. It does not tell you what those tokens accomplished. Across a company, millions of tokens are wasted as employees build similar agents for the same tasks. Sharing agents helps, but duplication is
- X
Wikis aren't enough: designing memory for multi-agent systems.
Wikis aren't enough: designing memory for multi-agent systems. Read more. https://t.co/45VGFxPlLM
- X
Humans hallucinate too...
Humans hallucinate too... Humans can also be prompt injected. Humans also have context windows. So why do we trust humans but not agents? Identity. We know who did what. Reputation compounds. Accountability sticks. That infrastructure doesn't exist for agents yet. Enforceable...
- X
The founder models organizational memory as a permissioned multidimensional graph and argues that agents will hold overlapping but distinct slices of company knowledge.
The founder models organizational memory as a permissioned multidimensional graph and argues that agents will hold overlapping but distinct slices of company knowledge.