How Pandas Helps You Land a QA Tester Job Faster

import pandas as pd

# test_runs.csv: run_date,module,test_name,status,duration_s
runs = pd.read_csv("test_runs.csv", parse_dates=["run_date"])
runs["failed"] = runs["status"].eq("failed")
by_module = (runs.groupby("module")["failed"]
                 .agg(runs="size", fail_rate="mean")
                 .sort_values("fail_rate", ascending=False))
print(by_module.assign(fail_rate=lambda d: (d["fail_rate"] * 100).round(1)))

weekly = (runs.set_index("run_date")
              .resample("W")["failed"]
              .mean()
              .rsub(1)            # pass rate = 1 - fail rate
              .mul(100)
              .round(1))
print(weekly.tail(8))
flaky = runs.groupby("test_name")["status"].nunique()
print("Tests with mixed results:", list(flaky[flaky > 1].index)[:10])

import pandas as pd

bugs = pd.read_csv("defects_export.csv", parse_dates=["created"])
open_bugs = bugs[bugs["status"].isin(["Open", "In Progress", "Reopened"])]
summary = pd.crosstab(open_bugs["component"], open_bugs["severity"], margins=True)
print(summary)

today = pd.Timestamp("2026-11-06")
open_bugs = open_bugs.assign(age_days=(today - open_bugs["created"]).dt.days)
stuck = (open_bugs[open_bugs["age_days"] > 14]
         .sort_values(["severity", "age_days"], ascending=[True, False])
         [["key", "summary", "severity", "component", "age_days"]])
print(stuck.head(15).to_string(index=False))

import pandas as pd

methods = pd.DataFrame({"method": ["UPI", "Card", "COD"]})
cities = pd.DataFrame({"city": ["Noida", "Delhi", "Ghaziabad"]})
amounts = pd.DataFrame({"amount": [99, 499, 5000]})
cases = methods.merge(cities, how="cross").merge(amounts, how="cross")
cases["expect_cod_allowed"] = ~((cases["method"] == "COD") & (cases["amount"] > 2000))
cases.to_csv("payment_cases.csv", index=False)
print(len(cases), "test cases written")

import pandas as pd

qa = pd.read_csv("prices_qa.csv")          # product_id,price,tax
stg = pd.read_csv("prices_staging.csv")
diff = qa.merge(stg, on="product_id", suffixes=("_qa", "_stg"))
mismatch = diff[(diff["price_qa"] != diff["price_stg"]) | (diff["tax_qa"] != diff["tax_stg"])]
print(len(mismatch), "products differ between QA and staging")
print(mismatch.head())

import pandas as pd

runs = pd.read_csv("test_runs.csv", parse_dates=["run_date"])
last_week = runs[runs["run_date"] >= runs["run_date"].max() - pd.Timedelta(days=7)]
report = {
    "test_runs": len(last_week),
    "pass_rate_%": round((last_week["status"] == "passed").mean() * 100, 1),
    "worst_module": last_week[last_week["status"] == "failed"]["module"].mode().iat[0]
                    if (last_week["status"] == "failed").any() else "none",
}
print(report)

# Use the generated CSV in pytest
import pandas as pd
import pytest

CASES = pd.read_csv("payment_cases.csv").to_dict("records")

@pytest.mark.parametrize("case", CASES, ids=lambda c: f"{c['method']}-{c['city']}-{c['amount']}")
def test_cod_rule(case):
    allowed = not (case["method"] == "COD" and case["amount"] > 2000)
    assert allowed == case["expect_cod_allowed"]

QA data checklist
[ ] Export includes dates, module/component, status
[ ] Status values consistent (passed/failed/skipped)
[ ] No personal data in exported fields
[ ] Scripts saved in the team repo
[ ] Weekly report shared with the same format each time