Problem Solving and Communication Questions

from typing import Iterable, List, Dict

def remove_duplicates(records: Iterable[Dict]) -> List[Dict]:
    """
    Keep only the latest record per 'id' based on 'timestamp' (ISO string).
    Assumes timestamps are ISO 8601 comparable lexicographically (e.g. "2023-05-01T12:00:00Z").
    Returns a list of latest records (order not guaranteed).
    """
    latest_per_id = {}  # id -> record

    for rec in records:
        rec_id = rec.get('id')
        ts = rec.get('timestamp')
        if rec_id is None or ts is None:
            # skip or handle malformed records; here we skip
            continue

        # Compare ISO strings directly (cheaper than parsing to datetime)
        existing = latest_per_id.get(rec_id)
        if existing is None or ts > existing['timestamp']:
            latest_per_id[rec_id] = rec

    return list(latest_per_id.values())

WITH ranked AS (
  SELECT event_id
  FROM (
    SELECT event_id,
           ROW_NUMBER() OVER (PARTITION BY user_id, event_type
                              ORDER BY occurred_at DESC, event_id DESC) AS rn
    FROM events
  ) t
  WHERE rn > 1
)
DELETE FROM events e
USING ranked r
WHERE e.event_id = r.event_id;