I made PostgreSQL INSERT operations 36 times faster in Python

Compares multiple approaches to bulk PostgreSQL INSERT operations in Python, showing how switching from pg8000 to psycopg2 and using `execute_values` instead of SQLAlchemy's executemany can yield up to 37x performance improvement. The key insight is that the default executemany approach sends each row as a separate round-trip, while `psycopg2.extras.execute_values` batches all rows into a single INSERT statement. Also covers a raw string VALUES construction approach with a SQL injection warning, and briefly mentions asyncpg for async workloads.