From 5e2dae13623b9849025bbdf333789c47c4d52c62 Mon Sep 17 00:00:00 2001
From: "google-labs-jules[bot]"
 <161369871+google-labs-jules[bot]@users.noreply.github.com>
Date: Fri, 23 May 2025 23:17:56 +0000
Subject: [PATCH] feat: Add emojis to README.md for better scannability

Adds topic-appropriate emojis to section headers and key list items
in README.md. This change aims to make the document more visually
engaging and easier to scan, based on your feedback.
---
 README.md | 79 +++++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 77 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index b0d9ea3..fc49a08 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,77 @@
-# gbatchkit-python
-GCP Batch Toolkit
+# ☁️🛠️ gbatchkit
+
+## 🎯 Overview
+
+`gbatchkit` is a Python library that simplifies defining, configuring, and submitting Google Cloud Batch jobs. It offers:
+
+*   **Programmatic Job Definition:** Construct Batch jobs in Python (machine types, GPUs, storage, etc.).
+*   **Task Argument Handling:** Parse arguments for batch tasks, with Pydantic support.
+*   **Simplified Submission:** A wrapper around `gcloud` for job submission.
+*   **Multitask Utilities:** Helpers for jobs with multiple runnables/arguments.
+
+**Typical Workflow:**
+
+1.  ➡️ **Define Task Args:** Use `gbatchkit.inputs.get_task_arguments` in your task script.
+2.  ➡️ **Build Job:** Use `gbatchkit.jobs` and `gbatchkit.types` (e.g., `create_standard_job`, `ContainerRunnable`) to create the job dict.
+3.  ➡️ **(Optional) Prep Multitask Args:** Use `jobs.prepare_multitask_job` if tasks need varied arguments.
+4.  ➡️ **Submit:** Call `jobs.submit_job`.
+
+Key modules: `gbatchkit.jobs`, `gbatchkit.types`, `gbatchkit.inputs`.
+
+## 🚀 Simple Usage Example
+
+Define and submit a basic Google Cloud Batch job:
+
+```python
+from gbatchkit import jobs, types
+
+# Define a container and compute resources
+container = types.ContainerRunnable(
+    image_uri="gcr.io/google-containers/busybox",
+    entrypoint="/bin/sh",
+    commands=["-c", "echo Hello from task ${BATCH_TASK_INDEX}! && sleep 10"]
+)
+compute_config = types.ComputeConfig(machine_type="n1-standard-1") # SPOT is default
+
+# Create job definition (single task)
+job_definition = jobs.create_standard_job(
+    region="us-central1", # Specify your GCP region
+    compute_config=compute_config,
+    task_count=1,
+    runnables=[container]
+)
+
+# For jobs with multiple tasks needing different arguments, `jobs.prepare_multitask_job`
+# serializes a list of argument dicts (one per task) to a specified cloud storage
+# `working_directory`. Each task then uses `inputs.get_task_arguments()` to load
+# its specific arguments from files within that directory (e.g., `task_0_args.json`).
+# The job definition is updated with environment variables pointing to these argument files.
+
+# Submit the job
+try:
+    job_id = "my-gbatchkit-job"
+    jobs.submit_job(job=job_definition, job_id=job_id, region="us-central1")
+    print(f"Submitted job: {job_id}")
+except RuntimeError as e:
+    print(f"Error: {e}")
+```
+
+## 📦 Installation
+
+Install `gbatchkit` from PyPI (recommended) or source:
+
+```bash
+pip install gbatchkit  # From PyPI
+pip install .          # From local source
+```
+Requires Python 3.9+. Dependencies are in `requirements.txt`.
+
+## ✅ Running Tests
+
+Tests use `pytest`. Install dev dependencies and run:
+
+```bash
+pip install .[dev]
+pytest
+```
+The tests are in the `tests/` directory.