Extend structured output

vacancies/main/__init__.py

@@ -0,0 +1,3 @@
+import logging
+
+logging.basicConfig(level=logging.INFO)

vacancies/main/bot.py

@@ -126,25 +126,24 @@ async def handle_document(update: Update, context: ContextTypes.DEFAULT_TYPE):
         job_titles: list[Literal[tuple(db_job_titles)]]
         min_salary_rub: int | None
         max_salary_rub: int | None
+        years_of_experience: int | None
+        work_format: Literal["Удаленный", "Офис", "Гибрид", None]
 
-    openai_client = ChatOpenAI(model_name="gpt-5-mini", temperature=0, seed=42, top_p=1)
+    openai_client = ChatOpenAI(model_name="gpt-5-mini")
     structured_llm = openai_client.with_structured_output(Structure)
 
-    prompt = f"""
-        Ты — HR-классификатор. Ниже приведён список допустимых профессий.
-        Твоя задача — выбрать наиболее подходящие по смыслу.
-        Качество классификации - самое важное.
-        Игнорируй орфографические и стилистические различия.
-        Резюме:
-        {resume}
-    """
+    prompt = f"Extract sturcture from following CV. {resume}"
+    print('1')
     response = await structured_llm.ainvoke(prompt)
+    print('2')
 
     customer = await Customer.objects.aget(telegram_id=update.effective_user.id)
     customer_cv, _ = await CustomerCV.objects.aupdate_or_create(customer=customer, defaults=dict(
         content=resume,
         min_salary_rub=response.min_salary_rub,
         max_salary_rub=response.max_salary_rub,
+        years_of_experience=response.years_of_experience,
+        work_format=response.work_format,
     ))
     await customer_cv.job_titles.aset([job_title_map[job_title] for job_title in response.job_titles])
 

vacancies/main/management/commands/collect_vacancies_from_telegram_messages.py

@@ -42,6 +42,8 @@ class Command(BaseCommand):
             job_title: Literal[tuple(job_titles)]
             min_salary_rub: int | None
             max_salary_rub: int | None
+            min_years_of_experience: int
+            work_format: Literal["remote", "office", None]
 
         openai_client = ChatOpenAI(model_name="gpt-5-mini", temperature=0, seed=42, top_p=1)
         structured_llm = openai_client.with_structured_output(Structure)
@@ -76,6 +78,8 @@ class Command(BaseCommand):
                     job_title_id=job_title_map[response.job_title],
                     min_salary_rub=response.min_salary_rub,
                     max_salary_rub=response.max_salary_rub,
+                    min_years_of_experience=response.min_years_of_experience,
+                    work_format=response.work_format,
                     content=message,
                     timestamp=timezone.make_aware(timestamp),
                     link=f"https://t.me/{chat_username}/{telegram_id}",

vacancies/main/migrations/0012_customercv_work_format_and_more.py

@@ -0,0 +1,35 @@
+# Generated by Django 5.2.7 on 2025-11-09 12:54
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('main', '0011_remove_customercv_job_title_customercv_job_titles_and_more'),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name='customercv',
+            name='work_format',
+            field=models.CharField(blank=True, max_length=64, null=True),
+        ),
+        migrations.AddField(
+            model_name='customercv',
+            name='years_of_experience',
+            field=models.PositiveBigIntegerField(default=0),
+            preserve_default=False,
+        ),
+        migrations.AddField(
+            model_name='vacancy',
+            name='min_years_of_experience',
+            field=models.PositiveBigIntegerField(default=0),
+            preserve_default=False,
+        ),
+        migrations.AddField(
+            model_name='vacancy',
+            name='work_format',
+            field=models.CharField(blank=True, max_length=64, null=True),
+        ),
+    ]

vacancies/main/models.py

@@ -28,6 +28,8 @@ class CustomerCV(models.Model):
     job_titles = models.ManyToManyField(JobTitle, related_name="vacancies")
     min_salary_rub = models.PositiveIntegerField(null=True, blank=True, default=None)
     max_salary_rub = models.PositiveIntegerField(null=True, blank=True, default=None)
+    years_of_experience = models.PositiveBigIntegerField()
+    work_format = models.CharField(max_length=64, null=True, blank=True)
     content = models.TextField()
     created_at = models.DateTimeField(auto_now_add=True)
 
@@ -45,6 +47,8 @@ class Vacancy(models.Model):
     external_id = models.CharField(max_length=255, unique=True)
     min_salary_rub = models.PositiveIntegerField(null=True, blank=True, default=None)
     max_salary_rub = models.PositiveIntegerField(null=True, blank=True, default=None)
+    min_years_of_experience = models.PositiveBigIntegerField()
+    work_format = models.CharField(max_length=64, null=True, blank=True)
     content = models.TextField()
     timestamp = models.DateTimeField()
     link = models.URLField()

vacancies/main/recommendations.py

@@ -6,7 +6,9 @@ def get_next_vacancy(customer_cv):
     vacancy = Vacancy.objects.filter(
         ~Q(id__in=customer_cv.customer.recommended_vacancies.values_list("vacancy_id", flat=True)),
         Q(min_salary_rub__isnull=True) | Q(min_salary_rub__gt=customer_cv.min_salary_rub),
+        Q(work_format__isnull=True) | Q(work_format=customer_cv.work_format),
         job_title__title__in=customer_cv.job_titles.values_list("title", flat=True),
+        min_years_of_experience__lte=customer_cv.years_of_experience,
     ).first()
     if vacancy:
         customer_cv.customer.recommended_vacancies.create(vacancy=vacancy)