Merge branch 'master' into readme

frontend/src/components/CustomHeader/index.tsx

@@ -64,7 +64,7 @@ export function CustomHeader() {
           <Title className={classes.modalHeader}>
             How does{" "}
             <Text component="span" className={classes.highlight} inherit>
-              Semantic search
+              Code search
             </Text>{" "}
             work?
           </Title>
@@ -80,23 +80,28 @@ export function CustomHeader() {
           }}
         >
           <Text size="lg" color="dimmed" className={classes.description}>
-            The search page will allow users to search for code snippets using
-            natural language. The text input will be converted into a vector
-            representation using advanced machine learning techniques. This
-            vector will then be used to semantically search a code snippet
-            database, retrieving similar code based on its meaning and
-            functionality.
+            On the search interface, it's possible for users to initiate searches
+            for code snippets by employing either natural language queries or
+            more code-like examples. The system processes the input by transforming
+            it into vector representations through the use of two neural encoders.
+            These vectors play a crucial role in semantically searching a database
+            filled with code snippets, ensuring the retrieval of code that aligns
+            closely with the intended meaning and/or functionality.
           </Text>
 
           <Image src="/workflow.svg" />
           <Text size="lg" color="dimmed" className={classes.description}>
-            The search results will display code snippets that are most relevant
-            to the user's query, ranked by their similarity to the input text.
-            Users can view and compare the retrieved code snippets to find the
-            one that best suits their needs. This approach to code search aims
-            to improve the efficiency and accuracy of finding relevant code by
-            leveraging advanced natural language processing and machine learning
-            algorithms.
+            The outcome of this search presents users with code snippets that closely
+            match their query, organized in order of their relevance to the entered
+            text. This allows users to explore and evaluate the found code snippets
+            to identify the one that meets their requirements best. By utilizing
+            natural language processing and machine learning techniques, this code
+            search method enhances the precision and efficiency in locating pertinent
+            code. Moreover, it facilitates the exploration of unfamiliar codebases
+            even without the necessity of prior knowledge in the programming language,
+            overcoming the limitations of traditional keyword-based searches which
+            falter without knowledge of specific variable or class names. Thus, semantic
+            code search significantly improves the search experience.
           </Text>
           <Button
             className={classes.modalBtnInner}