Events normalization job

Normalize raw event records into a clean list of analytics-ready dictionaries with stable field names and validation handling.

Separate normalization from validation so the main flow stays readable during discussion.

Sample input

raw event records

Mixed-quality event payloads from an analytics ingestion job.

[
  {
    "user_id": 42,
    "session_id": "s-100",
    "event_name": "page_view",
    "event_ts": "2026-03-20T09:00:00Z"
  },
  {
    "user_id": 42,
    "session_id": "s-100",
    "event_name": "add_to_cart",
    "event_ts": "2026-03-20T09:04:00Z"
  },
  {
    "user_id": 77,
    "session_id": "s-301",
    "event_name": "checkout_start",
    "event_ts": "bad_timestamp"
  }
]

Expected output

expected return shape

Keep normalized records and validation issues separate.

{
  "normalized": [
    {
      "user_id": 42,
      "session_id": "s-100",
      "event_name": "add_to_cart",
      "event_ts": "2026-03-20T09:04:00Z"
    }
  ],
  "errors": [
    {
      "row_index": 2,
      "reason": "malformed timestamp"
    }
  ]
}

Constraints

Malformed timestamps should be captured in an error summary. Preserve only the newest event per user-session pair.

Expected skills

Parsing, validation, edge-case handling, and writing transformation logic that is easy to review.

Python

AI evaluation

Run the transform to inspect normalized rows and error summaries.