Hi everyone, I'm Jean — a Total Rewards professional currently exploring the AI Safety space.
I've spent the last few years designing incentive systems for humans: compensation structures, recognition programs, performance frameworks. The core challenge is always the same — how do you translate an organization's values into measurable criteria, without creating perverse incentives that undermine the very goals you're trying to reach?
At some point I realized this is exactly the alignment problem.
The more I read about RLHF, reward hacking, and Constitutional AI, the more I recognized patterns I'd already encountered in my work. Goodhart's Law isn't a theoretical concern for me — I've watched it play out in bonus systems. The tension between extrinsic and intrinsic motivation? I've navigated that in practice.
I'm at the beginning of this transition. I've just started working through AI Safety Fundamentals and I'm looking to connect with people thinking seriously about the human side of alignment — governance, evaluation, values specification.
If you work at the intersection of behavioral systems and AI alignment, or if you think Total Rewards intuitions could translate meaningfully into this field, I'd genuinely love to talk.
Hi everyone, I'm Jean — a Total Rewards professional currently exploring the AI Safety space.
I've spent the last few years designing incentive systems for humans: compensation structures, recognition programs, performance frameworks. The core challenge is always the same — how do you translate an organization's values into measurable criteria, without creating perverse incentives that undermine the very goals you're trying to reach?
At some point I realized this is exactly the alignment problem.
The more I read about RLHF, reward hacking, and Constitutional AI, the more I recognized patterns I'd already encountered in my work. Goodhart's Law isn't a theoretical concern for me — I've watched it play out in bonus systems. The tension between extrinsic and intrinsic motivation? I've navigated that in practice.
I'm at the beginning of this transition. I've just started working through AI Safety Fundamentals and I'm looking to connect with people thinking seriously about the human side of alignment — governance, evaluation, values specification.
If you work at the intersection of behavioral systems and AI alignment, or if you think Total Rewards intuitions could translate meaningfully into this field, I'd genuinely love to talk.
Happy to be here.