In a scenario where a file is written into S3, how would you trigger an AWS Glue job that takes the file as an input argument? What steps would you follow?

When a file lands in an S3 bucket you can automatically start a Glue job by using an S3 event notification. 1) Configure the bucket to publish a PutObject event to an Amazon SNS topic or directly to an AWS Lambda function. 2) Create a Lambda function that receives the event, extracts the S3 key and bucket name, and calls the Glue API start_job_run passing the key as a job argument (e.g., --arguments "--input_path=s3://bucket/key"). 3) In the Glue job script, read the argument and use it to locate the file (e.g., via Spark spark.read.parquet(args['input_path'])). 4) Optionally wrap the Lambda and Glue call in an AWS Step Functions state machine for better error handling and retries. 5) Monitor the job via CloudWatch logs and Glue console. This pattern decouples the ingestion from the ETL logic and scales automatically.