more informative quickstart

Author: alexzhang13

examples/quickstart.py

@@ -1,4 +1,6 @@
 import os
+import random
+import string
 
 from dotenv import load_dotenv
 
@@ -7,21 +9,31 @@
 
 load_dotenv()
 
-logger = RLMLogger(log_dir="./logs")
+# Generate a large text file with a hidden secret number
+secret_number = random.randint(100_000_000, 999_999_999)
+filler_lines = ["".join(random.choices(string.ascii_lowercase + " ", k=120)) for _ in range(50_000)]
+insert_at = random.randint(len(filler_lines) // 3, 2 * len(filler_lines) // 3)
+filler_lines.insert(insert_at, f"SECRET_NUMBER={secret_number}")
+haystack = "\n".join(filler_lines)
 
 rlm = RLM(
-    backend="portkey",  # or "openai", etc.
+    backend="openai",
     backend_kwargs={
-        "model_name": "@openai/gpt-5-nano",
-        "api_key": os.getenv("PORTKEY_API_KEY"),
+        "model_name": "gpt-5-nano",
+        "api_key": os.getenv("OPENAI_API_KEY"),
     },
     environment="local",
-    environment_kwargs={},
-    max_depth=1,
-    logger=logger,
-    verbose=True,  # For printing to console with rich, disabled by default.
+    max_iterations=10,
+    logger=RLMLogger(log_dir="./logs"),
+    verbose=True,
 )
 
-result = rlm.completion("Print me the first 5 powers of two, each on a newline.")
+result = rlm.completion(
+    "The context contains ~50k lines of random text with a single line "
+    "matching the pattern SECRET_NUMBER=<digits>. Find and return ONLY the "
+    f"numeric value.\n\n{haystack}"
+)
 
-print(result)
+print(f"\nModel found: {result.response}")
+print(f"Actual number: {secret_number}")
+print(f"Correct: {str(secret_number) in result.response}")