Rag engine

Makefile

@@ -59,9 +59,15 @@ grpcui: ## Open gRPC UI in browser
 	@echo "Opening gRPC UI at http://localhost:8081"
 	@open http://localhost:8081 2>/dev/null || xdg-open http://localhost:8081 2>/dev/null || echo "Open http://localhost:8081 in your browser"
 
-db: ## Open SQLite Web browser
-	@echo "Opening SQLite Web at http://localhost:8082"
-	@open http://localhost:8082 2>/dev/null || xdg-open http://localhost:8082 2>/dev/null || echo "Open http://localhost:8082 in your browser"
+db: ## Open pgAdmin web interface (starts with --profile tools if needed)
+	@if ! docker ps --format '{{.Names}}' | grep -q fi-pgadmin; then \
+		echo "Starting pgAdmin..."; \
+		docker-compose --profile tools up -d pgadmin; \
+		sleep 2; \
+	fi
+	@echo "Opening pgAdmin at http://localhost:5050"
+	@echo "Login: admin@fiapply.local / admin"
+	@open http://localhost:5050 2>/dev/null || xdg-open http://localhost:5050 2>/dev/null || echo "Open http://localhost:5050 in your browser"
 
 # Status
 status: ## Show status of all services
@@ -92,5 +98,12 @@ quickstart: install build up ## Install, build, and start everything
 	@echo "Frontend:    http://localhost:3000"
 	@echo "Swagger UI:  http://localhost:8080"
 	@echo "gRPC UI:     http://localhost:8081"
-	@echo "DB Browser:  http://localhost:8082"
+	@echo "pgAdmin:     http://localhost:5050"
 	@echo ""
+
+PYTHONPATH := $(shell pwd)
+
+# Run pytest on the RAG engine tests
+test-rag:
+	@echo "🧪 Running RAG engine tests..."
+	PYTHONPATH=$(PYTHONPATH) pytest -s -v --full-trace backend/services/rag_engine/tests/test.py

README.md

@@ -12,7 +12,7 @@ FiApply/
 ├── backend/        # Python gRPC microservices
 ├── frontend/       # React TypeScript PWA
 ├── proto/          # Protocol Buffer definitions
-├── db/             # SQLite database
+├── db/             # PostgreSQL database
 └── docs/           # Documentation
 ```
 
@@ -25,7 +25,7 @@ FiApply/
 **Tech Stack:**
 - **Backend**: Python 3.11+, gRPC, Connect RPC, Playwright, LangChain, LangGraph
 - **Frontend**: React 18, TypeScript, Vite, ShadCN UI, Redux Toolkit, Connect-Web
-- **Database**: SQLite with vector extensions
+- **Database**: PostgreSQL 17-alpine3.22 with pgvector, pg_trgm, pg_stat_statements
 - **AI**: Stagehand (browser automation), Ollama/OpenRouter (LLMs)
 
 ---
@@ -63,7 +63,7 @@ python backend/services/llm_manager/example_client.py
 - 🎨 **Frontend**: http://localhost:3000
 - 📚 **API Docs**: http://localhost:8080 (Swagger UI)
 - 🔧 **gRPC Testing**: http://localhost:8081 (gRPC UI)
-- 💾 **Database**: http://localhost:8082 (SQLite Web)
+- 💾 **Database**: http://localhost:5050 (pgAdmin)
 
 See **[docs/DOCKER.md](docs/DOCKER.md)** for detailed Docker documentation.
 
@@ -285,16 +285,17 @@ All tools run automatically with Docker.
 
 | Service | Port | Description |
 |---------|------|-------------|
+| **postgres** | 5432 | PostgreSQL database |
+| **pgadmin** | 5050 | Database management UI |
 | **ollama** | 11434 | Local LLM runtime |
 | **swagger-ui** | 8080 | API documentation |
 | **grpcui** | 8081 | Interactive gRPC testing |
-| **sqlite-web** | 8082 | Database browser |
 
 Access via:
 ```zsh
 make swagger    # http://localhost:8080
 make grpcui     # http://localhost:8081
-make db         # http://localhost:8082
+make db         # http://localhost:5050 (pgAdmin)
 ```
 
 ---
@@ -325,7 +326,7 @@ FiApply/
 │   ├── ai_orchestrator.proto
 │   ├── rag_engine.proto
 │   └── llm_manager.proto
-├── db/                        # SQLite database
+├── db/                        # PostgreSQL database
 ├── docs/                      # Documentation
 ├── docker-compose.yml         # Full stack orchestration
 ├── Makefile                   # Development commands
@@ -390,7 +391,11 @@ STAGEHAND_API_KEY=...
 OPENROUTER_API_KEY=...
 
 # Database
-DB_PATH=/path/to/production.db
+DB_HOST=postgres
+DB_PORT=5432
+DB_NAME=fiapply
+DB_USER=fiapply_user
+DB_PASSWORD=fiapply_password
 
 # Service ports (optional)
 GATEWAY_PORT=50051
@@ -457,8 +462,8 @@ docker-compose exec ai-orchestrator playwright install chromium
 - ✅ OpenAPI/Swagger documentation
 - ✅ Browser automation with Stagehand
 - ✅ Connect RPC (gRPC + HTTP/JSON support)
+- ✅ PostgreSQL database with vector search (pgvector)
 - ⏳ Service implementations (see TODO comments in code)
-- ⏳ Database schema and migrations
 - ⏳ Authentication system
 - ⏳ Production deployment configs
 

backend/generated/ai_orchestrator_pb2_grpc.py

@@ -2,7 +2,7 @@
 """Client and server classes corresponding to protobuf-defined services."""
 import grpc
 
-from backend.generated import ai_orchestrator_pb2 as ai__orchestrator__pb2
+import ai_orchestrator_pb2 as ai__orchestrator__pb2
 
 
 class AIOrchestratorStub(object):

backend/generated/api_gateway_pb2_grpc.py

@@ -2,9 +2,9 @@
 """Client and server classes corresponding to protobuf-defined services."""
 import grpc
 
-from backend.generated import api_gateway_pb2 as api__gateway__pb2
+import api_gateway_pb2 as api__gateway__pb2
 from google.protobuf import empty_pb2 as google_dot_protobuf_dot_empty__pb2
-from backend.generated import headless_streaming_pb2 as headless__streaming__pb2
+import headless_streaming_pb2 as headless__streaming__pb2
 
 
 class GatewayStub(object):

backend/generated/llm_manager_pb2_grpc.py

@@ -2,7 +2,7 @@
 """Client and server classes corresponding to protobuf-defined services."""
 import grpc
 
-from backend.generated import llm_manager_pb2 as llm__manager__pb2
+import llm_manager_pb2 as llm__manager__pb2
 
 
 class LLMManagerStub(object):

backend/generated/rag_engine_pb2.pyi

@@ -106,3 +106,65 @@ class SearchResult(_message.Message):
     score: float
     metadata: _struct_pb2.Struct
     def __init__(self, document_id: _Optional[str] = ..., content: _Optional[str] = ..., score: _Optional[float] = ..., metadata: _Optional[_Union[_struct_pb2.Struct, _Mapping]] = ...) -> None: ...
+
+class InitRequest(_message.Message):
+    __slots__ = ("connection_string", "collection_name")
+    CONNECTION_STRING_FIELD_NUMBER: _ClassVar[int]
+    COLLECTION_NAME_FIELD_NUMBER: _ClassVar[int]
+    connection_string: str
+    collection_name: str
+    def __init__(self, connection_string: _Optional[str] = ..., collection_name: _Optional[str] = ...) -> None: ...
+
+class InitResponse(_message.Message):
+    __slots__ = ("success", "error")
+    SUCCESS_FIELD_NUMBER: _ClassVar[int]
+    ERROR_FIELD_NUMBER: _ClassVar[int]
+    success: bool
+    error: _common_pb2.Error
+    def __init__(self, success: _Optional[bool] = ..., error: _Optional[_Union[_common_pb2.Error, _Mapping]] = ...) -> None: ...
+
+class EmbedScholarshipBatchRequest(_message.Message):
+    __slots__ = ("scholarships",)
+    SCHOLARSHIPS_FIELD_NUMBER: _ClassVar[int]
+    scholarships: _containers.RepeatedCompositeFieldContainer[Scholarship]
+    def __init__(self, scholarships: _Optional[_Iterable[_Union[Scholarship, _Mapping]]] = ...) -> None: ...
+
+class EmbedScholarshipBatchResponse(_message.Message):
+    __slots__ = ("total_embedded", "error")
+    TOTAL_EMBEDDED_FIELD_NUMBER: _ClassVar[int]
+    ERROR_FIELD_NUMBER: _ClassVar[int]
+    total_embedded: int
+    error: _common_pb2.Error
+    def __init__(self, total_embedded: _Optional[int] = ..., error: _Optional[_Union[_common_pb2.Error, _Mapping]] = ...) -> None: ...
+
+class Document(_message.Message):
+    __slots__ = ("document_id", "content", "metadata")
+    DOCUMENT_ID_FIELD_NUMBER: _ClassVar[int]
+    CONTENT_FIELD_NUMBER: _ClassVar[int]
+    METADATA_FIELD_NUMBER: _ClassVar[int]
+    document_id: str
+    content: str
+    metadata: _struct_pb2.Struct
+    def __init__(self, document_id: _Optional[str] = ..., content: _Optional[str] = ..., metadata: _Optional[_Union[_struct_pb2.Struct, _Mapping]] = ...) -> None: ...
+
+class Scholarship(_message.Message):
+    __slots__ = ("id", "name", "description", "min_gpa", "eligible_majors", "citizenship_required", "location", "award_amount", "keywords")
+    ID_FIELD_NUMBER: _ClassVar[int]
+    NAME_FIELD_NUMBER: _ClassVar[int]
+    DESCRIPTION_FIELD_NUMBER: _ClassVar[int]
+    MIN_GPA_FIELD_NUMBER: _ClassVar[int]
+    ELIGIBLE_MAJORS_FIELD_NUMBER: _ClassVar[int]
+    CITIZENSHIP_REQUIRED_FIELD_NUMBER: _ClassVar[int]
+    LOCATION_FIELD_NUMBER: _ClassVar[int]
+    AWARD_AMOUNT_FIELD_NUMBER: _ClassVar[int]
+    KEYWORDS_FIELD_NUMBER: _ClassVar[int]
+    id: int
+    name: str
+    description: str
+    min_gpa: float
+    eligible_majors: _containers.RepeatedScalarFieldContainer[str]
+    citizenship_required: str
+    location: str
+    award_amount: str
+    keywords: _containers.RepeatedScalarFieldContainer[str]
+    def __init__(self, id: _Optional[int] = ..., name: _Optional[str] = ..., description: _Optional[str] = ..., min_gpa: _Optional[float] = ..., eligible_majors: _Optional[_Iterable[str]] = ..., citizenship_required: _Optional[str] = ..., location: _Optional[str] = ..., award_amount: _Optional[str] = ..., keywords: _Optional[_Iterable[str]] = ...) -> None: ...