diff --git a/scripts/similarity/config.yml b/scripts/similarity/config.yml
index 2f546488..1372d28d 100644
--- a/scripts/similarity/config.yml
+++ b/scripts/similarity/config.yml
@@ -2,5 +2,9 @@
 # note this is sensitive data
 # handle with care
 # DO NOT COMMIT THIS FILE WITH YOUR API KEY
-api_keys:
-    qdrant: YOUR_API_KEY
+
+cohere:
+    api_key: COHERE_API_KEY
+qdrant:
+    api_key: QDRANT_API_KEY
+    url: QDRANT_CLUSTER_URL
\ No newline at end of file
diff --git a/scripts/similarity/qdrant_search.py b/scripts/similarity/qdrant_search.py
deleted file mode 100644
index 6bfb298a..00000000
--- a/scripts/similarity/qdrant_search.py
+++ /dev/null
@@ -1,92 +0,0 @@
-import yaml
-from qdrant_client import QdrantClient, models
-from sentence_transformers import SentenceTransformer
-
-# NOTE: DO NOT RUN THIS FILE IN YOUR LOCAL SYSTEM UNLESS
-# YOU HAVE RAM > 16GB + NVIDIA GPU
-# SENTENCE_TRANSFORMERS HAVE DEPENDECY ON PYTORCH GPU
-# DON'T RUN UNLESS YOU HAVE THE CAPACITY
-# I'M WORKING ON A WAY TO USE THIS FILE IN COLAB
-
-
-class QdrantSearch:
-    def __init__(self, api_key, documents, query_string):
-        self.api_key = api_key
-        self.documents = documents
-        self.query_string = query_string
-        # This is take from the examples provided at Qdrant
-        self.url = "https://4248f066-5345-4bec-b3c3-ca343a34747e.us-east-1-0.aws.cloud.qdrant.io:6333"
-        # There is a disk option is also available.
-        self.qdrant = QdrantClient(":memory:")
-        # This needs to be downloaded.
-        self.encoder = SentenceTransformer('all-MiniLM-L6-v2')
-
-    def init_qdrant_client(self):
-        self.qdrant = QdrantClient(
-            url=self.url,
-            api_key=self.api_key,
-        )
-        self.qdrant.recreate_collection(
-            collection_name="resume_matcher",
-            vectors_config=models.VectorParams(
-                size=self.encoder.get_sentence_embedding_dimension(),
-                distance=models.Distance.COSINE
-            )
-        )
-        self.qdrant.upload_records(
-            collection_name="resume_matcher",
-            records=[
-                models.Record(
-                    id=idx,
-                    vector=self.encoder.encode(doc["resume"]).tolist(),
-                    payload=doc
-                ) for idx, doc in enumerate(self.documents)
-            ]
-        )
-
-    def search_documents(self):
-        """
-        Note the query string provided needs to be a job description.
-        """
-
-        if not self.qdrant:
-            self.init_qdrant_client()
-
-        hits = self.qdrant.search(
-            collection_name="resume_matcher",
-            query_vector=self.encoder.encode(self.query_string).tolist(),
-            limit=20
-        )
-
-        results = []
-        for hit in hits:
-            result = {
-                'text': str(hit.payload)[:30],
-                'query': self.query_string[:50],
-                'score': hit.score,
-            }
-            results.append(result)
-
-        return results
-
-
-# Get the API key stored in config.yml
-def read_api_key(filepath):
-    with open(filepath, 'r') as file:
-        try:
-            data = yaml.safe_load(file)
-            return data['api_keys']['qdrant']
-        except yaml.YAMLError as exc:
-            print(exc)
-
-
-api_key = read_api_key('qdrant-api-key.yml')
-
-
-# ### MAIN ###
-
-# output = QdrantSearch(api_key, resumes, job["job_desc"])
-# output.init_qdrant_client()
-# results = output.search_documents()
-# for result in results:
-#     print(result)