Upload job_matching.py

src/job_matching.py

@@ -0,0 +1,35 @@
+from sklearn.metrics.pairwise import cosine_similarity
+import numpy as np
+from chroma_utils import search_resume_chroma
+
+def calculate_ats_score(job_embedding):
+    
+    try:
+
+        search_results = search_resume_chroma(job_embedding, k=10)
+        candidate_ids = search_results['ids'][0]
+        candidate_embeddings = search_results['embeddings'][0]
+        candidate_metadatas = search_results['metadatas'][0]
+        
+        matched_candidates = []
+        for i, candidate_id in enumerate(candidate_ids):
+
+            candidate_metadata = candidate_metadatas[i]
+            candidate_embedding = np.array(candidate_embeddings[i])
+            
+            similarity_score = cosine_similarity([job_embedding], [candidate_embedding])[0][0]
+            
+            matched_candidates.append({
+                "candidate_id": candidate_id,
+                "score": similarity_score,
+                "metadata": candidate_metadata
+            })
+            
+        matched_candidates.sort(key=lambda x: x["score"], reverse=True)
+        return matched_candidates
+    except Exception:
+        raise
+
+
+    
+