Added lab 6 instructions

Author: siddhi

bruno/Jobify/Get Resume Recommendation.bru

@@ -0,0 +1,20 @@
+meta {
+  name: Get Resume Recommendation
+  type: http
+  seq: 12
+}
+
+get {
+  url: {{BASE_URL}}/api/job-posts/1/recommend
+  body: none
+  auth: inherit
+}
+
+vars:pre-request {
+  BASE_URL: http://127.0.0.1:8000
+}
+
+settings {
+  encodeUrl: true
+  timeout: 0
+}

labs/06-retrieval.md

@@ -0,0 +1,118 @@
+# Lab 6: Retrieve Recommended Resume
+
+In this lab, we will add an API endpoint to retrieve the recommended resume and return it.
+
+Here are the steps to complete:
+
+1. Create a function `get_recommendation` in `ai.py` that takes the job description
+1. Create an api endpoint `api_recommend_resume` in `main.py`
+    - Path `/api/job-posts/{job_post_id}/recommend`
+    - Takes `job_post_id` as a parameter
+    - Calls the `get_recommendation` function created in previous step
+    - Returns the matching job application
+
+Write tests along with the code.
+
+## High Level Overview
+
+1. We want to implement `get_recommendatation` in `ai.py`. First think about how we will test this function.
+1. One way can be to add some resumes to the vector store, then call `get_recommendation` with a job description and see whether it is returning the right one
+1. Write a test to do the above. It will fail when you run the tests, since we have not implemented it yet
+1. Make the test pass by implementing the functionality in `get_recommendation`
+1. Next we need to implement the API. Again -- how will we test this?
+1. Maybe we can create a dummy job post with a job description and then upload some resumes. Then we can call the recommendation API using the test client and see if we get the correct output.
+1. Write this a pyunit test that does this.
+1. Then implement the api endpoint till it passes
+1. We are done!
+
+## Hints
+
+### How do I test get_recommendatation?
+
+<details>
+<summary>Answer</summary>
+
+```python
+def test_retrieval(vector_store):
+    for id, filename in enumerate(os.listdir("test/resumes")):
+        with open(f"test/resumes/{filename}", "rb") as f:
+            content = f.read()
+        ingest_resume_for_recommendataions(content, filename, resume_id=id, vector_store=vector_store)
+    result = get_recommendation("I am looking for an expert in AI", vector_store)
+    assert "Andrew" in result.page_content
+```
+</details>
+
+### How do I implement get_recommendation?
+
+<details>
+<summary>Answer</summary>
+
+```python
+def get_recommendation(job_description, vector_store):
+    retriever = vector_store.as_retriever(search_kwargs={"k": 1})
+    results = retriever.invoke(job_description)
+    return results[0]
+```
+</details>
+
+### How do I test the new API endpoint?
+
+<details>
+<summary>Answer</summary>
+
+```python
+def test_recommendation_api(db_session, client):
+    job_board = JobBoard(slug="test", logo_url="http://example.com")
+    db_session.add(job_board)
+    db_session.commit()
+    db_session.refresh(job_board)
+    job_post = JobPost(title="AI Engineer", 
+                       description="I am looking for a generalist who can work in python and typescript", 
+                       job_board_id = job_board.id)
+    db_session.add(job_post)
+    db_session.commit()
+    db_session.refresh(job_post)
+    test_resumes = [
+        ('Andrew', 'Ng', '[email protected]', 'ProfileAndrewNg.pdf'),
+        ('Koudai', 'Aono', '[email protected]', 'ProfileKoudaiAono.pdf')
+    ]
+    for first_name, last_name, email, filename in test_resumes:
+        post_data = {
+            "first_name": first_name,
+            "last_name": last_name,
+            "email": email,
+            "job_post_id": job_post.id
+        }
+        with open(f"test/resumes/{filename}", "rb") as f:
+            response = client.post("/api/job-applications", data=post_data, files={"resume": (filename, f, "application/pdf")})
+    response = client.get(f"/api/job-posts/{job_post.id}/recommend")
+    assert response.status_code == 200
+    data = response.json()
+    assert data['first_name'] == 'Koudai'
+    assert data['last_name'] == 'Aono'
+```
+</details>
+
+### How do I implement the new API endpoint?
+
+<details>
+<summary>Answer</summary>
+
+```python
+@app.get("/api/job-posts/{job_post_id}/recommend")
+async def api_recommend_resume(
+   job_post_id, 
+   db: Session = Depends(get_db),
+   vector_store = Depends(get_vector_store)):
+   
+   job_post = db.get(JobPost, job_post_id)
+   if not job_post:
+      raise HTTPException(status_code=400)
+   job_description = job_post.description
+   recommended_resume = get_recommendation(job_description, vector_store)   
+   application_id = recommended_resume.metadata["_id"]
+   job_application = db.get(JobApplication, application_id)
+   return job_application
+```
+</details>

labs/07-homework.md

@@ -0,0 +1,19 @@
+# Lab 7: Homework
+
+Update the UI to use the new recommendataion system.
+
+Since this is a repeat of frontend / backend, we won't be doing this in the class. You can try it out by yourself as a practise exercise.
+
+## Part 1
+
+We will start by creating a new page to view job details. 
+
+1. Create a new route to view a single job. In the jobs board page, instead of showing apply button and job dscription, link the job title to route to this new page.
+    - What should be the URL structure for this new page?
+1. Create a backend api to return details of a single job. Input will be job id. Return the job details as well as the list of all applicants for the job - name, email, resume url
+1. Update the single job page to get the job id from url, and call this api. Display the job title, job description. Then show the apply button and then the list of all people who have applied for that job.
+
+## Part 2
+
+1. On this page, add a new button called "Get Recommendation". 
+1. When the button is clicked, make a call to the recommendations API that we created in previous lab. Display the applicant name and link to the resume on the screen

main.py

@@ -6,7 +6,7 @@
 from pydantic import BaseModel, EmailStr, Field
 from typing import List
 from sqlalchemy import text
-from ai import evaluate_resume_with_ai, review_application, ingest_resume, get_vector_store
+from ai import evaluate_resume_with_ai, get_recommendation, review_application, ingest_resume, get_vector_store
 from auth import AdminAuthzMiddleware, AdminSessionMiddleware, authenticate_admin, delete_admin_session
 from converter import extract_text_from_pdf_bytes
 from db import get_db
@@ -110,7 +110,22 @@ async def api_close_job_post(job_post_id, db: Session = Depends(get_db)):
    db.add(jobPost)
    db.commit()
    return jobPost
-  
+
+@app.get("/api/job-posts/{job_post_id}/recommend")
+async def api_recommend_resume(
+   job_post_id, 
+   db: Session = Depends(get_db),
+   vector_store = Depends(get_vector_store)):
+   
+   job_post = db.get(JobPost, job_post_id)
+   if not job_post:
+      raise HTTPException(status_code=400)
+   job_description = job_post.description
+   recommended_resume = get_recommendation(job_description, vector_store)   
+   application_id = recommended_resume.metadata["_id"]
+   job_application = db.get(JobApplication, application_id)
+   return job_application
+
 class JobPostForm(BaseModel):
    title : str
    description: str

test/test_recommendation.py

@@ -1,6 +1,6 @@
 import os
 import time
-from ai import ingest_resume
+from ai import get_recommendation, ingest_resume
 from main import ingest_resume_for_recommendataions
 from models import JobBoard, JobPost
 
@@ -65,16 +65,40 @@ def test_job_application_api(db_session, vector_store, client):
     assert "Andrew" in result[0].page_content
     assert result[0].metadata["_id"] == job_post.id
 
-"""
-text = "hello world"
-url = "helloworld.pdf"
-doc = Document(page_content=text, metadata={"file_url": url})
-embeddings = OpenAIEmbeddings(model="text-embedding-3-small", api_key=settings.OPENAI_API_KEY)
+def test_retrieval(vector_store):
+    for id, filename in enumerate(os.listdir("test/resumes")):
+        with open(f"test/resumes/{filename}", "rb") as f:
+            content = f.read()
+        ingest_resume_for_recommendataions(content, filename, resume_id=id, vector_store=vector_store)
+    result = get_recommendation("I am looking for an expert in AI", vector_store)
+    assert "Andrew" in result.page_content
 
-doc_store = QdrantVectorStore.from_documents(
-    [doc], embeddings, location=":memory:", collection_name="testing"
-)
-retriever = doc_store.as_retriever(search_kwargs={"k": 1})
-result = retriever.invoke("hello")
-print(result)
-"""
+def test_recommendation_api(db_session, client):
+    job_board = JobBoard(slug="test", logo_url="http://example.com")
+    db_session.add(job_board)
+    db_session.commit()
+    db_session.refresh(job_board)
+    job_post = JobPost(title="AI Engineer", 
+                       description="I am looking for a generalist who can work in python and typescript", 
+                       job_board_id = job_board.id)
+    db_session.add(job_post)
+    db_session.commit()
+    db_session.refresh(job_post)
+    test_resumes = [
+        ('Andrew', 'Ng', '[email protected]', 'ProfileAndrewNg.pdf'),
+        ('Koudai', 'Aono', '[email protected]', 'ProfileKoudaiAono.pdf')
+    ]
+    for first_name, last_name, email, filename in test_resumes:
+        post_data = {
+            "first_name": first_name,
+            "last_name": last_name,
+            "email": email,
+            "job_post_id": job_post.id
+        }
+        with open(f"test/resumes/{filename}", "rb") as f:
+            response = client.post("/api/job-applications", data=post_data, files={"resume": (filename, f, "application/pdf")})
+    response = client.get(f"/api/job-posts/{job_post.id}/recommend")
+    assert response.status_code == 200
+    data = response.json()
+    assert data['first_name'] == 'Koudai'
+    assert data['last_name'] == 'Aono'