Optimized Prompt Engineering and Tech Description

app/alembic/env.py

@@ -68,9 +68,7 @@ def run_migrations_online() -> None:
     )
 
     with connectable.connect() as connection:
-        context.configure(
-            connection=connection, target_metadata=target_metadata
-        )
+        context.configure(connection=connection, target_metadata=target_metadata)
 
         with context.begin_transaction():
             context.run_migrations()

app/alembic/versions/417469bac4de_removed_auth_type_from_otp.py

@@ -0,0 +1,40 @@
+"""removed auth-type from OTP
+
+Revision ID: 417469bac4de
+Revises: a6d1c6c13842
+Create Date: 2025-08-31 09:57:49.661025
+
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "417469bac4de"
+down_revision: Union[str, Sequence[str], None] = "a6d1c6c13842"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Upgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column("otps", "provider_id")
+    op.drop_column("otps", "auth_type")
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    """Downgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column(
+        "otps",
+        sa.Column("auth_type", sa.VARCHAR(), autoincrement=False, nullable=False),
+    )
+    op.add_column(
+        "otps",
+        sa.Column("provider_id", sa.VARCHAR(), autoincrement=False, nullable=True),
+    )
+    # ### end Alembic commands ###

app/alembic/versions/a4b5fabfdeac_add_otp_table.py

@@ -5,43 +5,45 @@
 Create Date: 2025-07-24 18:30:31.802426
 
 """
+
 from typing import Sequence, Union
 
 import sqlalchemy as sa
 from alembic import op
 
 # revision identifiers, used by Alembic.
-revision: str = 'a4b5fabfdeac'
-down_revision: Union[str, Sequence[str], None] = 'ccf5dc612027'
+revision: str = "a4b5fabfdeac"
+down_revision: Union[str, Sequence[str], None] = "ccf5dc612027"
 branch_labels: Union[str, Sequence[str], None] = None
 depends_on: Union[str, Sequence[str], None] = None
 
 
 def upgrade() -> None:
     """Upgrade schema."""
     # ### commands auto generated by Alembic - please adjust! ###
-    op.create_table('otps',
-    sa.Column('id', sa.Integer(), nullable=False),
-    sa.Column('email', sa.String(), nullable=False),
-    sa.Column('otp_code', sa.String(), nullable=False),
-    sa.Column('name', sa.String(), nullable=False),
-    sa.Column('password', sa.String(), nullable=True),
-    sa.Column('is_active', sa.Integer(), nullable=True),
-    sa.Column('provider_id', sa.String(), nullable=True),
-    sa.Column('auth_type', sa.String(), nullable=False),
-    sa.Column('created_at', sa.DateTime(), nullable=True),
-    sa.Column('expires_at', sa.DateTime(), nullable=False),
-    sa.PrimaryKeyConstraint('id')
+    op.create_table(
+        "otps",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("email", sa.String(), nullable=False),
+        sa.Column("otp_code", sa.String(), nullable=False),
+        sa.Column("name", sa.String(), nullable=False),
+        sa.Column("password", sa.String(), nullable=True),
+        sa.Column("is_active", sa.Integer(), nullable=True),
+        sa.Column("provider_id", sa.String(), nullable=True),
+        sa.Column("auth_type", sa.String(), nullable=False),
+        sa.Column("created_at", sa.DateTime(), nullable=True),
+        sa.Column("expires_at", sa.DateTime(), nullable=False),
+        sa.PrimaryKeyConstraint("id"),
     )
-    op.create_index(op.f('ix_otps_email'), 'otps', ['email'], unique=False)
-    op.create_index(op.f('ix_otps_id'), 'otps', ['id'], unique=False)
+    op.create_index(op.f("ix_otps_email"), "otps", ["email"], unique=False)
+    op.create_index(op.f("ix_otps_id"), "otps", ["id"], unique=False)
     # ### end Alembic commands ###
 
 
 def downgrade() -> None:
     """Downgrade schema."""
     # ### commands auto generated by Alembic - please adjust! ###
-    op.drop_index(op.f('ix_otps_id'), table_name='otps')
-    op.drop_index(op.f('ix_otps_email'), table_name='otps')
-    op.drop_table('otps')
+    op.drop_index(op.f("ix_otps_id"), table_name="otps")
+    op.drop_index(op.f("ix_otps_email"), table_name="otps")
+    op.drop_table("otps")
     # ### end Alembic commands ###

app/alembic/versions/a6d1c6c13842_add_role_field.py

@@ -0,0 +1,38 @@
+"""add role field
+
+Revision ID: a6d1c6c13842
+Revises: a4b5fabfdeac
+Create Date: 2025-08-31 09:15:05.419403
+
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = "a6d1c6c13842"
+down_revision: Union[str, Sequence[str], None] = "a4b5fabfdeac"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Upgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    userrole_enum = sa.Enum("ADMIN", "USER", name="userrole")
+    userrole_enum.create(op.get_bind())
+
+    # Then add the column using the enum type
+    op.add_column("users", sa.Column("role", userrole_enum, nullable=False))
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    """Downgrade schema."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column("users", "role")
+    # Then drop the enum type
+    sa.Enum(name="userrole").drop(op.get_bind())
+    # ### end Alembic commands ###

app/api/main.py

@@ -5,4 +5,6 @@
 api_router = APIRouter()
 
 # TODO: Add routes as needed
-api_router.include_router(questions.router, prefix="/questions", tags=["question_generation"])
+api_router.include_router(
+    questions.router, prefix="/questions", tags=["question_generation"]
+)

app/api/routes/questions.py

@@ -1,9 +1,10 @@
+import json
+
 from fastapi import APIRouter
 
-from app.schemas.utils import UserPrompt
-from app.services.questions import add_daily_question_to_mongodb
-from app.services.questions import \
-    add_daily_question_to_store as add_question_in_chroma_db
+from app.schemas.utils import TechDescription, UserPrompt
+from app.services.questions import (add_daily_question_to_mongodb,
+                                    add_tech_description_to_store)
 from app.services.questions import \
     generate_daily_question as generate_daily_question_service
 from app.services.questions import get_most_recent_daily_question
@@ -18,9 +19,26 @@ def generate_daily_question(request: UserPrompt):
     or user's input.
     """
     response = generate_daily_question_service(user_prompt=request.user_prompt)
-    add_daily_question_to_mongodb(response)  # Store question in MongoDB
+    try:
+        question = json.loads(response)  # Parse the JSON string to a dictionary
+    except json.JSONDecodeError:
+        return {"error": "Failed to parse the generated question."}
+    add_daily_question_to_mongodb(question)  # Store question in MongoDB
     return response
 
+
+@router.post("/add_tech_description")
+def add_tech_description(request: TechDescription):
+    """
+    Add a technical description to the vector store.
+    Whenever admin add a new tech description/Tag, it will be stored in ChromaDB.
+    """
+    add_tech_description_to_store(
+        description=request.description, metadata=request.metadata
+    )
+    return {"message": "Tech description added successfully."}
+
+
 @router.get("/get_daily_question")
 def get_daily_question():
     """
@@ -29,4 +47,4 @@ def get_daily_question():
     recent_question = get_most_recent_daily_question()
     if recent_question:
         return recent_question
-    return {"message": "No daily question found."}
+    return {"message": "No daily question found."}

app/config/config_mongo_fb.py

@@ -13,6 +13,7 @@ def __init__(self, uri: str, db_name: str):
     def get_collection(self, collection_name: str):
         return self.db[collection_name]
 
+
 # MongoDB connection setup
 MONGO_URI = "mongodb://mongodb:27017"
 MONGO_DB_NAME = "questions_db"

app/config/vectorestore.py

@@ -11,13 +11,14 @@
 # Using singleton pattern for ChromaDB to ensure a single instance is used across the application
 class ChromaDB:
     """Singleton class for managing ChromaDB collections and operations."""
-    _instance = None  
+
+    _instance = None
 
     def __new__(cls, *args, **kwargs):
         if cls._instance is None:
             cls._instance = super().__new__(cls)
         return cls._instance
-    
+
     def __init__(self):
         self.path = settings.CHROMA_PATH
         self.model = settings.EMBEDDING_MODEL
@@ -26,7 +27,7 @@ def __init__(self):
         self._client = PersistentClient(path=self.path)
 
         self._collections: Dict[str, Chroma] = {}
-    
+
     def get_collection(self, name: str) -> Chroma:
         """Get or create a Chroma-backed collection."""
         if name not in self._collections:
@@ -36,21 +37,23 @@ def get_collection(self, name: str) -> Chroma:
                 embedding_function=self._embeddings,
             )
         return self._collections[name]
-    
+
     def add_documents(self, collection: str, documents: List[Document]) -> None:
         """
         Add documents to the ChromaDB collection.
         """
         store = self.get_collection(collection)
         store.add_documents(documents)
-
-    def as_retriever(self,collection: str, search_kwargs: Optional[Dict[str, Any]] = None) -> Any:
+
+    def as_retriever(
+        self, collection: str, search_kwargs: Optional[Dict[str, Any]] = None
+    ) -> Any:
         """
         Get a retriever for the ChromaDB collection.
         search_kwargs: can be used to specify additional search parameters like 'k' for KNN.
         """
         store = self.get_collection(collection)
         return store.as_retriever(search_kwargs=search_kwargs)
-
 
-chroma_db = ChromaDB()
+
+chroma_db = ChromaDB()

app/llm/provider.py

@@ -17,7 +17,8 @@ def build_chat_model() -> ChatHuggingFace:
     )
     return ChatHuggingFace(llm=llm)
 
-#TODO: Improve this function
+
+# TODO: Improve this function
 def invoke_with_retries(chain, payload: Dict[str, Any]):
     last_err = None
     for attempt in range(1, settings.LLM_RETRIES + 1):
@@ -26,5 +27,7 @@ def invoke_with_retries(chain, payload: Dict[str, Any]):
         except Exception as e:  # noqa: BLE001
             last_err = e
             # Exponential backoff with cap
-            time.sleep(min(2 ** attempt, 8))
-    raise RuntimeError(f"LLM invocation failed after {settings.LLM_RETRIES} attempts") from last_err
+            time.sleep(min(2**attempt, 8))
+    raise RuntimeError(
+        f"LLM invocation failed after {settings.LLM_RETRIES} attempts"
+    ) from last_err

app/main.py

@@ -7,21 +7,13 @@
 app = FastAPI(
     title="Generator Service",
     description="API for generating and managing coding problems",
-    version="0.1.0"
-)
-
-# CORS middleware configuration
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"],  # In production, replace with specific origins
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
+    version="0.1.0",
 )
 
 app.include_router(api_router, prefix=settings.API_V1_STR, tags=["v1"])
 
+
 # Health check endpoint
 @app.get("/health")
 async def health_check():
-    return {"status": "healthy", "service": "JamAndFlow-generator"}
+    return {"status": "healthy", "service": "JamAndFlow-generator"}

app/models/__init__.py

@@ -1,3 +1,3 @@
 # Add models here to be imported by Alembic
 # example:
-# from .models.user import User # noqa: F401, F403
+# from .models.user import User # noqa: F401, F403

app/pipelines/daily_question_pipeline.py

@@ -12,17 +12,21 @@
 
 # using tech_description collection to retrieve context.
 # It will return top k relevant document based on the user prompt.
-# The context structure stores tech-stack and relevant real-world concepts 
+# The context structure stores tech-stack and relevant real-world concepts
 # On which the question can be framed.
-retriever = chroma_db.as_retriever(collection="tech_description",
-                                   search_kwargs={"k": settings.RETRIEVE_K})
+retriever = chroma_db.as_retriever(
+    collection="tech_description", search_kwargs={"k": settings.RETRIEVE_K}
+)
+
+prompt = ChatPromptTemplate.from_messages(
+    [
+        SystemMessagePromptTemplate.from_template(system_template),
+        HumanMessagePromptTemplate.from_template(human_template),
+    ]
+)
 
-prompt = ChatPromptTemplate.from_messages([
-    SystemMessagePromptTemplate.from_template(system_template),
-    HumanMessagePromptTemplate.from_template(human_template),
-])
 
-#TODO: we might required a context for last n question for uniquness and variety
+# TODO: we might required a context for last n question for uniquness and variety
 # But for now the concern is on the limit of the tokens hence not implementing it
 def _retrieve_context(inputs: Dict):
     """It will only trigger when user gives a contet if not it will return 'No extra context'"""
@@ -33,9 +37,9 @@ def _retrieve_context(inputs: Dict):
     context = " ".join([d.page_content for d in docs]) if docs else "No extra context"
     return {"context": context, "user_prompt": inputs["user_prompt"]}
 
+
 context_retriever = RunnableParallel(
-    user_question=lambda x: x["user_prompt"],
-    context=_retrieve_context
+    user_question=lambda x: x["user_prompt"], context=_retrieve_context
 )
 
 chat_model = build_chat_model()