Screen CVs from a Drive folder into Airtable

import io
from pypdf import PdfReader
from google.oauth2.credentials import Credentials
from googleapiclient.discovery import build
from clickruntime import on_trigger, credentials, http, ai, notify, state, report_fetched
BASE_ID, FOLDER_ID, JOB_RECORD = "appXXXXXXXXXXXXXX", "1FolderIdOfApplicants", "recXXXXXXXXXXXXXX"
SCHEMA = {"type": "object", "properties": {"name": {"type": "string"}, "email": {"type": "string"}, "skills": {"type": "string"},
          "score": {"type": "integer"}, "reason": {"type": "string"}}, "required": ["name", "score", "reason"]}
@on_trigger
def run(event):
    drive = build("drive", "v3", credentials=Credentials(credentials.connection("google", "work")), cache_discovery=False)
    since = state.get("since") or "2026-01-01T00:00:00Z"
    files = drive.files().list(q=f"'{FOLDER_ID}' in parents and createdTime > '{since}' and trashed = false",
                               orderBy="createdTime", fields="files(id,name,createdTime,webViewLink)").execute()["files"]
    report_fetched(len(files))
    if not files:
        return {"new": 0}
    api, headers = f"https://api.airtable.com/v0/{BASE_ID}", {"Authorization": f"Bearer {credentials.connection('airtable', 'work')}"}
    job = http.get(f"{api}/Jobs/{JOB_RECORD}", headers=headers).json()["fields"].get("Description", "")
    for f in files:
        text = "\n".join(p.extract_text() or "" for p in PdfReader(io.BytesIO(drive.files().get_media(fileId=f["id"]).execute())).pages)[:8000]
        c = ai.generate(f"Job:\n{job[:3000]}\n\nCV:\n{text}", json_schema=SCHEMA, max_output_tokens=250,
                        system="Extract name, email and key skills, score the CV 1-10 against the job, give a one-line reason.").json
        http.post(f"{api}/Candidates", headers=headers, json={"typecast": True, "records": [{"fields": {
            "Name": c["name"], "Email": c.get("email", ""), "Skills": c.get("skills", ""), "Score": c["score"],
            "Reason": c["reason"], "CV": f["webViewLink"]}}]}).raise_for_status()
        if c["score"] >= 8:
            notify.email_me(f"Strong candidate: {c['name']} ({c['score']}/10)", f"{c['reason']}\n{f['webViewLink']}")
        state.set("since", f["createdTime"])
    return {"new": len(files)}