The loop
- the user adds signal through onboarding, edits, uploads, tracked actions, or wins
- the profile spine becomes richer
- surfaces read that profile with the current mode layered on top
- the user gets a better next action, fit read, resume, or intelligence surface
- new movement creates more context for the next cycle
What counts as signal
school, cohort, years out, role, company, preferences, and current state
projects, work history, transcript facts, certifications, and skills
saved jobs, applications, plan usage, wins, and system interactions
target roles, career concerns, current loops, and where the user wants to move next
Example loop for a seeker
A seeker updates their profile, saves a target role, applies to three jobs, and tailors one resume in Forge. That should affect the jobs feed, home nudges, tracker urgency, and the next chat context. The system should understand that this is no longer a generic seeker. It is a seeker with active motion and narrowing intent.
Example loop for a holder
A holder adds their current company and role, starts reading Field Intelligence, and logs a win. That should influence weekly briefings, market comparisons, AI threat framing, and what “next move” means for that user.
What breaks the loop
- surfaces that ignore prior context
- artifacts that do not write useful evidence back into the system
- mode changes that behave like resets
- generic nudges that do not reflect tracked movement
- systems that over-collect but do not use what they already know well
Design consequence
Every new feature should answer one question clearly: what new signal does it create, and where does that signal become useful next?