Twitxy Saved

Conjecture

Verified account This account is verified. Learn more

@ConjectureAI

We are building Cognitive Emulation, an alternative AI paradigm that is controllable, transparent, and safe. Get in touch at hello@conjecture.dev.
London conjecture.dev Joined June 2022
4 Following 4K Followers 152 Posts

Latest posts 5

Pinned post

Conjecture Verified account @ConjectureAI
🚀Tactics just leveled up! We've added new features that dramatically expand what you can build. Think Firebase, but purpose-built for AI development. Easier, faster, more powerful. Your AI projects just got superpowered.

Come join our alpha test at alpha.conjecture.dev!

Among other features, you can now
- Use "Workspaces" to group tactics, manage state, and build full AI-enabled applications.
- Call OpenAI and Anthropic models.
- Call models with custom configurations, including JSON schema-constrained generation.
- Version control for better iterations on tactics.
- Explore a number of new example tactics that show off the range of capabilities tactics offers.

Full new features list is shared on our discord: discord.gg/AjZKAURk9f
5 replies 8 reposts 19 likes 8.4K views View on X
Conjecture Verified account @ConjectureAI
Tactics changelog from last week:

- We released tactics.dev/games as an example of how to build games with tactics, and an associated blog. Come try building your own game!

- We've also released a call for design partners. If you're enjoying tactics and want to get closer to the team to help us shape direction, let us know! tactics.dev/design-partners

- We updated access configuration to better control over who can access and run Tactics. Now, Tactics can be gated to only enable logged-in users to call them.

- CTAC now has type expressions (enums, objects, tuples) that enable you to ensure that the input and output of your Tactics conform to the right specification!

- Speaking of types, this may be a good time to mention that schema-following has been built into our LLM calls from the start! Write a JSON schema and add it to an LLM call, and you can guarantee the model will return in the desired format.