Optimizing the Unspoken Corporate Objective Function
You are good at games. You understand systems.
So why does your performance review feel like gaslighting?
KarmaRank is a framework for understanding how you are actually evaluated in modern corporations. Not the HR story, but the real objective function.
Stream in terminal:
curl -sL https://jimbojw.github.io/karmarank/karmarank-manifesto.7 | man -l -
Latest builds:
See Releases for stable versions.
Corporate performance evaluation is not a function of work quality; it is a function of Status-Weighted Stories.
This document patches the Default Engineering Worldview (which assumes meritocracy) with KarmaRank, a transitive, narrative-aggregating algorithm over the corporate social graph.
KarmaRank: Time-Decayed, Capital-Constrained, Adversarial Eigenvector-Sponsorship.
While this full KarmaRank definition describes the actual appraisal process, for the day-to-day practitioner, it suffices to optimize for Net Status-Weighted Karma:
Where:
-
$K$ — Net Status-Weighted Karma
In a given field: a company, a team, a subculture, a community. -
$i$ — The Index
Ranging over the people whose opinions about you actually travel—your manager, your manager's peers, the Staff engineer everyone listens to, the PM who won't shut up in calibration meetings. -
$S_i$ — Status
Intentionally capitalized, because it's the most important term. It's not just the witness's job title, but their actual ability to move decisions and narratives as it pertains to you. -
$k_i$ — Karmic Appraisal
The signed "emotional value" person$i$ gets from telling a story about you and your work.
The KarmaRank Manifesto proceeds from this observation, explaining how to decouple your self-worth from your rating (Law #1), optimize your output for the Firm Ledger, and maximize your karmic ROI within your local Calibration Cylinder.

Figure: Calibration Cylinder showing You (Defendant), your Manager (Counsel), your Manager's Peers (Prosecution/Jury) and your Skip Manager (Judge).
- License: CC-BY-NC-SA-4.0
- TL;DR
- Introduction: The Frame of the Game
- KarmaRank Origin Story (with Math)
- SECURITY ADVISORY: Five Vulnerabilities in the Default Engineering Worldview
- Law #1: Conative Dissonance
- Law #2: The Rating Is The Job
- Law #3: Who Cares?
- Meta-Law: You Do Not Talk About KarmaRank
- The Calibration Cylinder: Topology of the Trial
- Know Your Counsel: Technician, Politician, Peacemaker
- Tactics I: Karma-Driven Development (KDD)
- Tactics II: Manufacturing Evidence (The Brief)
- The Exit: The Market Check
- Conclusion: You Are Not Cynical Enough
- Epilogue: The Meta-Game
- Appendix
Contributions welcomed:
- 💬 GitHub Discussions for: Questions, Stories, Ideas, Debate.
- 🐛 GitHub Issues for actionable tasks only: Typos & Grammatical Errors, Rendering Bugs, Specific Content Requests.
See CONTRIBUTING.md for PR process details.
Requires Docker:
make -j$(nproc) all
Artifacts will be written to output/. See DEVELOPING.md for details.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
See the LICENSE file for the full legal text.

