feat: scaffold FastAPI app with async Postgres and Qdrant connection

backend/alembic.ini

@@ -0,0 +1,41 @@
+[alembic]
+script_location = alembic
+prepend_sys_path = .
+version_path_separator = os
+sqlalchemy.url = driver://user:pass@localhost/dbname
+
+[post_write_hooks]
+
+[loggers]
+keys = root,sqlalchemy,alembic
+
+[handlers]
+keys = console
+
+[formatters]
+keys = generic
+
+[logger_root]
+level = WARN
+handlers = console
+qualname =
+
+[logger_sqlalchemy]
+level = WARN
+handlers =
+qualname = sqlalchemy.engine
+
+[logger_alembic]
+level = INFO
+handlers =
+qualname = alembic
+
+[handler_console]
+class = StreamHandler
+args = (sys.stderr,)
+level = NOTSET
+formatter = generic
+
+[formatter_generic]
+format = %(levelname)-5.5s [%(name)s] %(message)s
+datefmt = %H:%M:%S

backend/alembic/README

@@ -0,0 +1 @@
+Generic single-database configuration.

backend/alembic/env.py

@@ -0,0 +1,61 @@
+import asyncio
+import re
+from logging.config import fileConfig
+from sqlalchemy import pool
+from sqlalchemy.ext.asyncio import create_async_engine
+from alembic import context
+
+config = context.config
+if config.config_file_name is not None:
+    fileConfig(config.config_file_name)
+
+import sys, os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
+
+from src.database import Base
+from src.models import JobDescription, Candidate, MatchResult
+from src.config import get_settings
+
+target_metadata = Base.metadata
+
+
+def _make_async_url(url: str) -> str:
+    url = re.sub(r"^postgresql:", "postgresql+asyncpg:", url)
+    url = re.sub(r"[?&]channel_binding=require", "", url)
+    return url
+
+
+def run_migrations_offline() -> None:
+    settings = get_settings()
+    context.configure(
+        url=_make_async_url(settings.database_url),
+        target_metadata=target_metadata,
+        literal_binds=True,
+        dialect_opts={"paramstyle": "named"},
+    )
+    with context.begin_transaction():
+        context.run_migrations()
+
+
+def do_run_migrations(connection):
+    context.configure(connection=connection, target_metadata=target_metadata)
+    with context.begin_transaction():
+        context.run_migrations()
+
+
+async def run_async_migrations() -> None:
+    settings = get_settings()
+    connectable = create_async_engine(_make_async_url(settings.database_url), poolclass=pool.NullPool)
+    async with connectable.connect() as connection:
+        await connection.run_sync(do_run_migrations)
+    await connectable.dispose()
+
+
+def run_migrations_online() -> None:
+    asyncio.run(run_async_migrations())
+
+
+if context.is_offline_mode():
+    run_migrations_offline()
+else:
+    run_migrations_online()

backend/alembic/script.py.mako

@@ -0,0 +1,23 @@
+"""${message}
+
+Revision ID: ${up_revision}
+Revises: ${down_revision | comma,n}
+Create Date: ${create_date}
+"""
+from typing import Sequence, Union
+from alembic import op
+import sqlalchemy as sa
+${imports if imports else ""}
+
+revision: str = ${repr(up_revision)}
+down_revision: Union[str, None] = ${repr(down_revision)}
+branch_labels: Union[str, Sequence[str], None] = ${repr(branch_labels)}
+depends_on: Union[str, Sequence[str], None] = ${repr(depends_on)}
+
+
+def upgrade() -> None:
+    ${upgrades if upgrades else "pass"}
+
+
+def downgrade() -> None:
+    ${downgrades if downgrades else "pass"}

backend/alembic/versions/001_initial_schema.py

@@ -0,0 +1,101 @@
+"""initial schema
+
+Revision ID: 001
+Revises:
+Create Date: 2026-04-11 10:00:00.000000
+"""
+from typing import Sequence, Union
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects.postgresql import UUID, JSON
+
+revision: str = "001"
+down_revision: Union[str, None] = None
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "job_descriptions",
+        sa.Column("id", UUID(as_uuid=True), primary_key=True),
+        sa.Column("title", sa.String(255), nullable=False),
+        sa.Column("raw_text", sa.Text, nullable=False),
+        sa.Column("parsed_requirements", JSON, nullable=True),
+        sa.Column("required_skills", JSON, nullable=True),
+        sa.Column("min_yoe", sa.Float, nullable=True),
+        sa.Column("max_yoe", sa.Float, nullable=True),
+        sa.Column("role_type", sa.String(100), nullable=True),
+        sa.Column("engineer_type", sa.String(100), nullable=True),
+        sa.Column("location", sa.String(255), nullable=True),
+        sa.Column("remote_allowed", sa.Boolean, nullable=True),
+        sa.Column("jd_quality", JSON, nullable=True),
+        sa.Column("embedding_text", sa.Text, nullable=True),
+        sa.Column("qdrant_id", sa.String(64), nullable=True),
+        sa.Column("status", sa.String(32), nullable=False, server_default="pending"),
+        sa.Column("created_at", sa.DateTime(timezone=True), server_default=sa.func.now()),
+        sa.Column("updated_at", sa.DateTime(timezone=True), server_default=sa.func.now()),
+    )
+    op.create_table(
+        "candidates",
+        sa.Column("id", UUID(as_uuid=True), primary_key=True),
+        sa.Column("external_id", sa.String(128), nullable=True),
+        sa.Column("name", sa.String(255), nullable=True),
+        sa.Column("email", sa.String(255), nullable=True),
+        sa.Column("looking_for", sa.String(100), nullable=True),
+        sa.Column("currently_employed", sa.Boolean, nullable=True),
+        sa.Column("notice_period", sa.String(100), nullable=True),
+        sa.Column("open_to_working_at", sa.String(255), nullable=True),
+        sa.Column("role_type", sa.String(100), nullable=True),
+        sa.Column("engineer_type", sa.String(100), nullable=True),
+        sa.Column("years_of_experience", sa.Float, nullable=True),
+        sa.Column("programming_languages", JSON, nullable=True),
+        sa.Column("backend_frameworks", JSON, nullable=True),
+        sa.Column("frontend_technologies", JSON, nullable=True),
+        sa.Column("gen_ai_experience", sa.Boolean, nullable=True),
+        sa.Column("recent_experience_type", sa.String(100), nullable=True),
+        sa.Column("education_status", sa.String(100), nullable=True),
+        sa.Column("degree", sa.String(255), nullable=True),
+        sa.Column("parsed_summary", sa.Text, nullable=True),
+        sa.Column("parsed_skills", sa.Text, nullable=True),
+        sa.Column("parsed_work_experience", JSON, nullable=True),
+        sa.Column("most_recent_company", sa.String(255), nullable=True),
+        sa.Column("most_recent_company_description", sa.Text, nullable=True),
+        sa.Column("most_recent_company_is_funded", sa.Boolean, nullable=True),
+        sa.Column("most_recent_company_is_product_company", sa.Boolean, nullable=True),
+        sa.Column("most_recent_company_total_funding", sa.Float, nullable=True),
+        sa.Column("most_recent_company_funding_status", sa.String(100), nullable=True),
+        sa.Column("time_in_current_company", sa.Float, nullable=True),
+        sa.Column("is_actively_or_passively_looking", sa.String(100), nullable=True),
+        sa.Column("growth_velocity", sa.Float, nullable=False, server_default="0.5"),
+        sa.Column("embedding_hash", sa.String(64), nullable=True),
+        sa.Column("qdrant_id", sa.String(64), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), server_default=sa.func.now()),
+        sa.Column("updated_at", sa.DateTime(timezone=True), server_default=sa.func.now()),
+    )
+    op.create_index("ix_candidates_external_id", "candidates", ["external_id"])
+    op.create_index("ix_candidates_qdrant_id", "candidates", ["qdrant_id"])
+    op.create_table(
+        "match_results",
+        sa.Column("id", UUID(as_uuid=True), primary_key=True),
+        sa.Column("jd_id", UUID(as_uuid=True), sa.ForeignKey("job_descriptions.id", ondelete="CASCADE"), nullable=False),
+        sa.Column("candidate_id", UUID(as_uuid=True), sa.ForeignKey("candidates.id", ondelete="CASCADE"), nullable=False),
+        sa.Column("rank", sa.Integer, nullable=True),
+        sa.Column("stage1_score", sa.Float, nullable=False, server_default="0"),
+        sa.Column("stage2_score", sa.Float, nullable=True),
+        sa.Column("final_score", sa.Float, nullable=False, server_default="0"),
+        sa.Column("component_scores", JSON, nullable=True),
+        sa.Column("gaps", JSON, nullable=True),
+        sa.Column("explanation", sa.Text, nullable=True),
+        sa.Column("explanation_generated_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("weights_used", JSON, nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), server_default=sa.func.now()),
+    )
+    op.create_index("ix_match_results_jd_id", "match_results", ["jd_id"])
+    op.create_index("ix_match_results_candidate_id", "match_results", ["candidate_id"])
+
+
+def downgrade() -> None:
+    op.drop_table("match_results")
+    op.drop_table("candidates")
+    op.drop_table("job_descriptions")

backend/main.py

@@ -0,0 +1,61 @@
+import os
+from contextlib import asynccontextmanager
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from qdrant_client import QdrantClient
+from qdrant_client.models import Distance, VectorParams
+
+from src.config import get_settings
+from src.database import engine
+from src.models import JobDescription, Candidate, MatchResult
+from src.routers import jds, candidates, matching
+
+settings = get_settings()
+
+_qdrant_client: QdrantClient | None = None
+
+
+def get_qdrant() -> QdrantClient:
+    return _qdrant_client
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    global _qdrant_client
+    _qdrant_client = QdrantClient(url=settings.qdrant_url, api_key=settings.qdrant_api_key)
+
+    existing = [c.name for c in _qdrant_client.get_collections().collections]
+    if settings.collection_name not in existing:
+        _qdrant_client.create_collection(
+            collection_name=settings.collection_name,
+            vectors_config=VectorParams(size=settings.vector_size, distance=Distance.COSINE),
+        )
+
+    app.state.qdrant = _qdrant_client
+    yield
+    _qdrant_client.close()
+
+
+app = FastAPI(
+    title="TalentPulse — AI Candidate Matching",
+    description="Two-stage retrieval + reranking pipeline for matching JDs against 100K+ candidates",
+    version="1.0.0",
+    lifespan=lifespan,
+)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+app.include_router(jds.router, prefix="/api/jds", tags=["Job Descriptions"])
+app.include_router(candidates.router, prefix="/api/candidates", tags=["Candidates"])
+app.include_router(matching.router, prefix="/api/match", tags=["Matching"])
+
+
+@app.get("/health")
+async def health():
+    return {"status": "ok", "version": "1.0.0"}

backend/requirements.txt

@@ -0,0 +1,20 @@
+fastapi==0.115.0
+uvicorn[standard]==0.30.6
+sqlalchemy[asyncio]==2.0.36
+asyncpg==0.30.0
+psycopg[binary]==3.2.3
+alembic==1.13.3
+qdrant-client==1.11.0
+celery[redis]==5.4.0
+redis==5.1.0
+sentence-transformers==3.3.0
+FlagEmbedding==1.2.15
+groq==0.12.0
+python-multipart==0.0.12
+pydantic==2.9.2
+pydantic-settings==2.5.2
+pandas==2.2.3
+numpy==1.26.4
+python-dotenv==1.0.1
+httpx==0.27.2
+aiofiles==24.1.0

backend/src/__init__.py

@@ -0,0 +1 @@
+# src package

backend/src/config.py

@@ -0,0 +1,22 @@
+from pydantic_settings import BaseSettings, SettingsConfigDict
+from functools import lru_cache
+
+
+class Settings(BaseSettings):
+    model_config = SettingsConfigDict(env_file=".env", env_file_encoding="utf-8", extra="ignore")
+
+    database_url: str
+    qdrant_url: str
+    qdrant_api_key: str
+    redis_url: str
+    groq_api_key: str
+    groq_model: str = "llama-3.3-70b-versatile"
+    embedding_model: str = "BAAI/bge-small-en-v1.5"
+    reranker_model: str = "BAAI/bge-reranker-v2-m3"
+    collection_name: str = "candidates_v1"
+    vector_size: int = 384
+
+
+@lru_cache
+def get_settings() -> Settings:
+    return Settings()

backend/src/database.py

@@ -0,0 +1,25 @@
+import re
+from typing import AsyncGenerator
+from sqlalchemy.ext.asyncio import create_async_engine, AsyncSession, async_sessionmaker
+from sqlalchemy.orm import DeclarativeBase
+from .config import get_settings
+
+
+def _make_async_url(url: str) -> str:
+    url = re.sub(r"^postgresql:", "postgresql+asyncpg:", url)
+    url = re.sub(r"[?&]channel_binding=require", "", url)
+    return url
+
+
+settings = get_settings()
+engine = create_async_engine(_make_async_url(settings.database_url), echo=False, pool_pre_ping=True)
+AsyncSessionLocal = async_sessionmaker(engine, expire_on_commit=False)
+
+
+class Base(DeclarativeBase):
+    pass
+
+
+async def get_db() -> AsyncGenerator[AsyncSession, None]:
+    async with AsyncSessionLocal() as session:
+        yield session