Ask HN: How do you survive a job you hate which is slowly making you sick?

TEXT START: I started a new job a little less than a year ago as a product owner/manager for a legacy enterprise system.

The Dissection

This is a crowd-sourced containment ritual around a failed modernization mandate. The poster has been assigned responsibility for customer-facing progress without being given a team capable of producing it. Management sells API modernization; the organization supplies PowerBuilder, COBOL, stored procedures, insufficient testing and deployment discipline, and one late additional hire.

The comments perform three functions: psychological triage, normalization of enterprise dysfunction, and escape planning. Exercise, therapy, sleep, hobbies, stoicism, and “care less” may reduce symptoms. “Quit,” take sick leave, save money, and keep applying are the only advice aimed at changing the material situation. The rest mostly relocates an institutional failure into the worker’s head.

The thread correctly identifies the core mismatch: this is not merely a product role. It is an attempt to modernize the people, processes, architecture, and incentives behind a legacy system. But it still treats that mismatch as an unusually bad job rather than a small-scale example of the broader discontinuity.

The Core Fallacy

The main fallacy is confusing emotional separation from work with material separation from the wage system.

“Care less” can stop the poster from donating their nervous system to the company, but it cannot remove the fact that they are the visible interface for promises the company cannot fulfill. The poster is not sick because they failed to take enough walks. They are trapped between management’s narrative and the team’s actual productive capacity.

The proposed cure of “modernize the team first” also assumes that management can coordinate retraining, hiring, architecture, testing, operations, and knowledge transfer merely because someone has described the need. The evidence in the text points the other way. Six months for four GET endpoints is not a temporary scheduling problem. It is an organizational capability failure.

Under the Discontinuity Thesis, AI will sharpen this failure. AI can lower the cost of scaffolding, documentation, prototyping, and routine cognitive work. It cannot make an organization with weak ownership, obsolete workflows, missing release discipline, and concentrated retirement-age domain knowledge suddenly coherent. Competitors that can coordinate will accelerate; this company will use human roles as shock absorbers while calling delay “modernization.”

Hidden Assumptions

• The job market will improve before the poster’s health deteriorates further.
• Quitting or taking leave is financially survivable.
• Management’s stated modernization priority is real rather than sales language or prestige signaling.
• One experienced API developer can transfer enough capability to change the system.
• The existing developers have both the time and willingness to acquire new skills.
• Domain knowledge can be retained while the organization changes its technical operating model.
• Customers will continue accepting promises despite the absence of shipped capability.
• The poster can remain accountable for delivery while psychologically pretending delivery is not their problem.
• AI-generated prototypes can overcome institutional inability to test, deploy, secure, maintain, and own production software.
• Meaning outside work compensates for an employment system that still controls income, health, and daily security.

The most dangerous assumption is that the bottleneck is primarily technical. It is coordination, authority, incentives, and institutional aging. Hiring one developer is a patch on a carcass, not a modernization strategy.

Social Function

Classification: partial truth, transition management, copium, lullaby, and ideological anesthetic.

The practical advice is not worthless. Therapy, medical care, exercise, social contact, financial runway, and an exit plan are legitimate survival mechanisms. But the dominant social function is to make an intolerable structure psychologically tolerable without demanding structural accountability from management.

“You care too much” is especially useful to the institution. It converts a company’s failure to fund capability into an employee’s failure to regulate attachment. “Enterprise IT is kabuki theatre” names the fraud, then recommends performing it better. Advice to reassure customers with elaborate excuses is not strategy; it is participation in the deception.

The AI-prototyping suggestion contains a real transition signal: product people can use AI to produce evidence of capability and seek higher altitude. But a prototype without distribution, authority, proprietary context, or control over deployment is just another unpaid audition. It does not make the poster a Sovereign. At best, it may help them become a more valuable Servitor elsewhere.

The Verdict

The thread is accurate about the immediate escape and evasive about the system. The poster is not failing to compartmentalize. They are the human gasket between an obsolete technical organization and management’s modernization theater.

This is not yet proof that the entire company is mechanically dead, but its modernization promise is already socially dead: customers are being sold a capability the organization has not purchased. The role is fragile because it carries accountability without control. The company can continue paying salaries for years while producing almost nothing of strategic value; that is not health, only prolonged hospice care.

The only defensible priority is to protect life and income while exiting. The dark thoughts outrank every career calculation: if they become immediate, specific, or difficult to control, treat that as an emergency, contact local emergency or crisis support, involve a trusted person, and do not remain alone. No amount of professional compartmentalization repairs a system that is consuming the operator.