Persist — The Process Continuity Platform

The Immortality Manifesto

Every process you've ever lost was a failure of infrastructure, not imagination.

For decades, the industry accepted a brutal truth: when the terminal dies, the process dies with it. A dropped SSH connection. A laptop lid closing at the wrong moment. A network timeout at 3am during a critical migration. And everything you were running — gone. We refused to accept that outcome. We built Persist because we believe no process should ever be subject to the mortality of its parent session. Your workloads are your intellectual property, your competitive advantage, your business. They deserve continuity that outlasts any single connection.

This is not incremental improvement. This is a fundamental shift in how the world thinks about process lifecycle. Immunity is not a feature — it's a right.

— The Persist Team

Capabilities

Six pillars of
process immortality.

Every feature engineered from first principles to ensure your workloads never die.

🛡️

SIGHUP Immunity Engine

Our core technology intercepts hangup signals before they reach your process. When the terminal disconnects, when SSH drops, when the session ends — your workload keeps running. Immune. Untouchable. Persistent.

persist ./deploy.sh

📋

Automatic Output Capture

If your process writes to a terminal, Persist seamlessly redirects stdout to a dedicated output file — nohup.out. Zero configuration. Every byte of output preserved, every line recoverable. No more lost logs.

persist ./migrate.sh
# output → nohup.out

📎

Smart Append Mode

When the output file already exists, Persist doesn't overwrite — it appends. Your historical logs stay intact, your audit trail grows, and compliance teams sleep at night. Every run's output compounds into a single source of truth.

persist ./batch-3.sh
# appends to existing nohup.out

🔀

Unified Error Consolidation

Standard error automatically redirects to standard output — giving you a single, unified stream of all process output. No more hunting through separate log channels. One file. Complete observability.

persist ./risky-job.sh 2>&1

🏠

Fallback Path Resolution

If the current directory can't accept output, Persist gracefully falls back to $HOME/nohup.out. Your process never fails because of a permission issue. Intelligent path resolution, handled automatically.

# CWD not writable → ~/nohup.out

📊

Exit Code Transparency

Persist preserves the exact exit code of your command — status 125 for internal failures, 126 for invocation errors, 127 for missing commands, and the original exit status otherwise. Full diagnostic fidelity.

# exit 0 in → exit 0 out

Architecture

Three steps to
process immortality.

A radically simple pipeline engineered for zero-downtime continuity.

Wrap Your Command

Prefix any command with persist. That's the entire integration. No SDK, no daemon, no YAML. One word transforms a mortal process into an immortal one.

Disconnect Freely

Close your terminal. Shut your laptop. Let your SSH session timeout. Persist intercepts the SIGHUP signal and shields your process from termination. It's still running.

Retrieve Your Output

Reconnect whenever you want and check nohup.out. Every line of stdout and stderr from your process is there, appended chronologically. Complete continuity.

Case Study

How Meridian Labs
eliminated process loss.

Meridian Labs Cloud Infrastructure

"We lost a 14-hour database migration because someone's VPN dropped. With Persist, that can never happen again."

Meridian Labs runs large-scale data migrations across multi-region PostgreSQL clusters. Their operations team regularly kicks off migration scripts that take anywhere from 4 to 18 hours. Before Persist, they had a catastrophic failure rate of nearly 12% — not because the scripts failed, but because the SSH sessions running them dropped due to network timeouts, laptop sleep events, or VPN reconnections.

After adopting Persist across their entire infrastructure team, Meridian Labs achieved a 100% migration completion rate. Every long-running job is now wrapped with Persist, making it immune to the session disconnections that plagued their operations. Output is automatically captured to nohup.out, giving them a complete audit trail of every migration.

The result: zero process losses, a 12% improvement in operational reliability, and an engineering team that finally trusts their infrastructure to run overnight without babysitting.

100% Migration completion rate

12%→0% Process loss rate

340+ Hours saved per quarter

Pricing

Continuity at every scale.

From solo operators to enterprise infrastructure teams.

Starter

Free

For individuals exploring process continuity

5 protected processes per day
Basic SIGHUP immunity
Output to nohup.out only
Community support

Get Started

Trusted by teams
that can't afford downtime.

"We used to schedule our long-running ETL jobs only when we knew someone could babysit the SSH session. With Persist, we just fire and forget. The process survives everything. It's a category-defining product."

Raj Krishnamurthy VP of Data Engineering, ScaleForge

"The automatic output capture alone is worth the subscription. We had a 22-hour model training run, our session dropped, but every single line of output was in nohup.out when we reconnected. Zero data loss."

Jamie Lindstrom ML Platform Lead, DeepGrid

"The stderr consolidation feature is incredibly underrated. One output stream, one file, complete observability. Our on-call engineers spend 60% less time triaging overnight job failures."

Felix Chen SRE Manager, CloudNova

"I pitched Persist to our CTO as 'process immortality as a service.' She approved the budget in the same meeting. It's that easy to explain and that hard to live without."

Lina Morales Staff DevOps Engineer, Alchemy IC

Team

The team behind
process immortality.

Rena Reeves

CEO & Co-founder

Former SRE lead at Google. Lost a 3-day MapReduce job to a dropped SSH session. Decided no one should ever experience that again.

Jason Kim

CTO & Co-founder

Signal processing PhD turned infrastructure engineer. Designed the SIGHUP interception layer from first principles. Obsessed with uptime.

Jenna Kirchhoff

Head of Reliability

Built the output capture and append pipeline. Previously led the process lifecycle team at Datadog. Believes every byte of output is sacred.

FAQ

Frequently asked questions.

What happens to my process when I close my terminal?

Without Persist, your process receives a SIGHUP signal and terminates. With Persist, that signal is intercepted and discarded. Your process continues running as if nothing happened. It's immune.

Where does the output go?

If your process would normally write to the terminal, Persist automatically redirects stdout to nohup.out in your current directory. If that directory isn't writable, it falls back to $HOME/nohup.out. Standard error is redirected to standard output, so everything ends up in one place.

Does it overwrite the output file if it already exists?

Never. Persist uses smart append mode — if nohup.out already exists, new output is appended to the end. Your historical data is always preserved. This is critical for compliance and audit trail requirements.

Can I redirect output to a custom file?

Absolutely. Use standard redirection: persist ./script.sh > custom-output.log. Persist respects your output configuration while still providing SIGHUP immunity. Available on Professional tier and above.

How do I know my process's exit status?

Persist provides full exit code transparency. If the command succeeds, you get its actual exit code. Status 125 indicates a Persist internal error, 126 means the command was found but couldn't be invoked, and 127 means the command wasn't found. Enterprise-grade diagnostics, built in.

Is my data secure?

Persist is SOC 2 Type II certified, ISO 27001 compliant, and GDPR ready. Output files are written with standard filesystem permissions. Enterprise customers get HIPAA-eligible configurations and on-premise deployment options.

Your processes
deserve immortality.

The Immortality Manifesto