# `Snakepit.WorkerProfile.Process`
[🔗](https://github.com/nshkrdotcom/snakepit/blob/v0.13.0/lib/snakepit/worker_profile/process.ex#L1)

Multi-process worker profile (default).

Each worker is a separate OS process, providing:
- **Process isolation**: Crashes don't affect other workers
- **GIL compatibility**: Works with all Python versions
- **High concurrency**: Optimal for 100+ workers with I/O-bound tasks

This is the default profile and maintains 100% backward compatibility
with Snakepit v0.5.x configurations.

## Configuration

    config :snakepit,
      pools: [
        %{
          name: :default,
          worker_profile: :process,  # Explicit (or omit for default)
          pool_size: 100,
          adapter_module: Snakepit.Adapters.GRPCPython,
          adapter_env: [
            {"OPENBLAS_NUM_THREADS", "1"},
            {"OMP_NUM_THREADS", "1"}
          ]
        }
      ]

## Implementation Details

- Each worker runs a single-threaded Python process
- Workers are single-capacity (one request at a time)
- Environment variables enforce single-threading in scientific libraries
- Startup is batched to prevent resource exhaustion

---

*Consult [api-reference.md](api-reference.md) for complete listing*