The base reasoning behind the exploratory work I’ve done with Aileron and shareable, portable, skills. Reproducible where appropriate and auditable always
Not everything is for inference
I predict the teams that best solve the deterministic / non-deterministic boundary will beat the teams throwing 100% of their work at models and agents.
Just because a transformer can:
- plan
- choose tools
- decide what to do next
- keep reasoning
- format outputs
- route between actions
doesn’t mean you should ask it to do all those things every time.
Much of that work may be better expressed as deterministic code for your use case.
@nachunja@TIDAL And, on its merits, the most high quality streaming service on the market.
Got fitted for IEMs and the audiologist team runs Tidal exclusively.
Corollary: Companies locking their data behind their own web apps with no APIs are going to be disrupted by challengers.
The special fancy data is the moat until your UX is so poor it invites others to jump in with the alligators
companies building their own agent rather than exposing an mcp / cli for their uses is like if you had to download a separate browser for every website you use