DuckDB version: v0.10.2,
Python version: 3.11.4

When I am trying to use the DATE_ADD function it raises a ParserException.
To recreate the issue,

data = [['2020-01-01', 4], ['2020-01-02',5]]
p = pd.DataFrame(data, columns=['cdate', 'days'])
print(p.info())
q = """
    SELECT DATE_ADD(CAST(p.cdate AS DATE), INTERVAL p.days DAY) pp
    FROM p 
"""
duckdb.sql(q).df()

which results in

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 2 entries, 0 to 1
Data columns (total 2 columns):
 #   Column  Non-Null Count  Dtype 
---  ------  --------------  ----- 
 0   cdate   2 non-null      object
 1   days    2 non-null      int64 
dtypes: int64(1), object(1)
memory usage: 164.0+ bytes
None
---------------------------------------------------------------------------
ParserException                           Traceback (most recent call last)
Cell In[42], line 8
      3 print(p.info())
      4 q = """
      5     SELECT DATE_ADD(CAST(p.cdate AS DATE), INTERVAL p.days DAY) pp
      6     FROM p 
      7 """
----> 8 duckdb.sql(q).df()

ParserException: Parser Error: syntax error at or near "p"

If I change the line

SELECT DATE_ADD(CAST(p.cdate AS DATE), INTERVAL p.days DAY) pp

to

SELECT DATE_ADD(CAST(p.cdate AS DATE), INTERVAL 5 DAY) pp

it is working as expected. Tried an explicit casting of p.days to INTEGER, which also led to the same exception.