Synthetic Users
@syntheticusers
Followers
678
Following
944
Media
111
Statuses
874
We’re building the future of research.
Inside Large Language Models
Joined November 2021
AI doesn’t replace human ingenuity — it amplifies it ⚡ @Ugo_alves, co-founder of @syntheticusers, returns for @lisbonaiweek to explore how intelligent agents are shifting creation, coordination & entrepreneurship back into human hands 🤝 Join his talk: https://t.co/6PVBfqx4NN
0
5
5
Hay una forma de entrevistar usuarios sin preguntar. Suena a trampa, pero no lo es 👀 Durante años, entender a los usuarios ha sido un proceso lento, caro y limitado. Y aun así, muchas decisiones se toman sobre suposiciones.
1
2
4
De esto vino a hablarnos @Ugo_alves, cofundador de @syntheticusers, en nuestra última Disciplina Interna de Beyond AB. Su herramienta crea “usuarios sintéticos”: perfiles generados por IA capaces de mantener entrevistas reales y ofrecer distintas perspectivas.
1
1
1
IMO — biggest misunderstanding in AI right now... Synthetic data isn't synthetic data. It's a cleaner version of the original data. It was named poorly.
59
20
415
We have personalities since day 0.
Giving LLM agents simple, psychology based personalities makes them more effective and easier to steer. Thinking agents defect in about 90% of Prisoner’s Dilemma rounds, Feeling agents about 50%. The paper’s recipe is MBTI-in-Thoughts, it assigns an MBTI type with a prompt
0
0
1
@Ugo_alves Watch my full conversation with Hugo Alves about the future of AI and Product Design:
1
2
2
“AI is just autocomplete.” Sure—but to autocomplete meaningfully, it has to model how humans think, act & aspire. That’s not trivial—it’s a map of human nature itself. More than math, it’s a mirror. #AI #innovation
1
2
3
What happens when AI doesn’t just support product design—but starts driving it? @Ugo_Alves, founder of Synthetic Users, sees a future where agent-based systems can design & prototype new products. So what role will humans play in that future? #productdesign #AI #innovation
1
2
4
I was skeptical about synthetic users—felt too far from real people. But Hugo Alves shared how a kombucha brand used them to test flavor ideas. The AI results lined up closely with real interviews. Takeaway: AI can speed the sort, humans seal the truth. #syntheticusers
2
2
3
Skeptical about synthetic users? Hear how a kombucha brand used them to test ideas fast: https://t.co/oa6ox6OKKk
0
2
1
0
1
1
It’s stuff like this that makes us. It’s amazing human insight out there in the open.
Fun satire of opposition to innovation, from 1833, which imagines geese petitioning against steel pens!
0
0
0
Grateful to have our recent ICML paper covered by @StanfordHAI. Humanity is building incredibly powerful AI technology that can usher in utopia or dystopia. We need human-computer interaction research, particularly with “social science tools like simulations that can keep pace.”
Social science research can be time-consuming, expensive, and hard to replicate. But with AI, scientists can now simulate human data and run studies at scale. Does it actually work?
0
2
6
whats it called when a business is named literally the thing it does
22
2
27
This
@chamath For those who don’t think they should build before they validate 👉 @syntheticusers
0
0
1
Can LLMs simulate human research subjects for psychology, economics, and other fields? Today I'm headed to Vancouver for #ICML2025 to present our position paper arguing yes! We *need* AI simulations so humanity's social understanding can keep pace with technological acceleration.
Should we use LLMs 🤖 to simulate human research subjects 🧑? In our new preprint, we argue sims can augment human studies to scale up social science as AI technology accelerates. We identify five tractable challenges and argue this is a promising and underused research method 🧵
0
2
4
Seems relevant.
When magnetic compasses came to Europe in the 12th c., ship captains had to hide them from sailors who by and large thought magnets were a form of witchcraft and refused to sail in their presence. Ship captain, sailor, and magnet are always with us in new and different forms.
0
0
0
Our original name was Synthz. @Ugo_alves originally wanted to go for Replicants (not dystopian at all) but was afraid of lawsuits.
0
1
3
Simulate, design, build, test, repeat Design, build, test, repeat, simulate Build, test, repeat, simulate, design Test, repeat, simulate, design, build repeat, simulate, design, build, test
8
11
97
We have a confession to make… We don’t use AI to generate our synthetic users. It’s Soham doing all the interviews.
PSA: there’s a guy named Soham Parekh (in India) who works at 3-4 startups at the same time. He’s been preying on YC companies and more. Beware. I fired this guy in his first week and told him to stop lying / scamming people. He hasn’t stopped a year later. No more excuses.
0
1
5