intro docker compose for ai eng

docker-compose-ai-eng/solution-files/.dockerignore

@@ -0,0 +1,2 @@
+.env
+__pycache__

docker-compose-ai-eng/solution-files/Dockerfile

@@ -0,0 +1,10 @@
+FROM python:3.10-slim
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+CMD ["uvicorn", "main:app", "--port", "8000", "--host", "0.0.0.0"]

docker-compose-ai-eng/solution-files/compose.yaml

@@ -0,0 +1,30 @@
+services:
+  app:
+    build: .
+    container_name: optimizer
+    ports:
+      - "8000:8000"
+    env_file:
+      - .env
+    environment:
+      DB_HOST: db
+      POSTGRES_DB: optimized_prompts
+      POSTGRES_USER: postgres
+      POSTGRES_PASSWORD: postgres
+    depends_on:
+      - db
+
+  db:
+    image: postgres:15
+    container_name: optimizer-db
+    environment:
+      POSTGRES_USER: postgres
+      POSTGRES_PASSWORD: postgres
+      POSTGRES_DB: optimized_prompts
+    ports:
+      - "5432:5432"
+    volumes:
+      - pgdata:/var/lib/postgresql/data
+
+volumes:
+  pgdata:

docker-compose-ai-eng/solution-files/database.py

@@ -0,0 +1,66 @@
+import psycopg2
+import os
+import time
+import logging
+from datetime import datetime, timezone
+
+logger = logging.getLogger(__name__)
+
+
+def get_db_connection():
+    """Create a connection to the PostgreSQL database."""
+    return psycopg2.connect(
+        host=os.getenv("DB_HOST", "db"),
+        port=int(os.getenv("DB_PORT", "5432")),
+        dbname=os.getenv("POSTGRES_DB", "optimized_prompts"),
+        user=os.getenv("POSTGRES_USER", "postgres"),
+        password=os.getenv("POSTGRES_PASSWORD", "postgres"),
+    )
+
+
+def init_db(max_retries=5, retry_delay=2):
+    """Create the optimization_logs table if it doesn't exist."""
+    for attempt in range(1, max_retries + 1):
+        try:
+            conn = get_db_connection()
+            cur = conn.cursor()
+            cur.execute("""
+                CREATE TABLE IF NOT EXISTS optimization_logs (
+                    id SERIAL PRIMARY KEY,
+                    original_prompt TEXT NOT NULL,
+                    optimized_prompt TEXT NOT NULL,
+                    changes TEXT NOT NULL,
+                    created_at TIMESTAMP NOT NULL DEFAULT NOW()
+                );
+            """)
+            conn.commit()
+            cur.close()
+            conn.close()
+            logger.info("Database initialized successfully.")
+            return
+        except psycopg2.OperationalError as e:
+            if attempt < max_retries:
+                logger.warning(
+                    f"Database not ready (attempt {attempt}/{max_retries}). "
+                    f"Retrying in {retry_delay}s..."
+                )
+                time.sleep(retry_delay)
+            else:
+                raise
+
+
+def log_optimization(original_prompt, optimized_prompt, changes):
+    """Write an optimization result to the database."""
+    conn = get_db_connection()
+    cur = conn.cursor()
+    cur.execute(
+        """
+        INSERT INTO optimization_logs
+            (original_prompt, optimized_prompt, changes, created_at)
+        VALUES (%s, %s, %s, %s);
+        """,
+        (original_prompt, optimized_prompt, changes, datetime.now(timezone.utc)),
+    )
+    conn.commit()
+    cur.close()
+    conn.close()

docker-compose-ai-eng/solution-files/main.py

@@ -0,0 +1,60 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel, Field
+from service import optimize_prompt
+import logging
+from database import init_db, log_optimization
+
+logger = logging.getLogger(__name__)
+
+app = FastAPI()
+
+init_db()
+
+@app.get("/health")
+def health_check():
+    return {"status": "ok"}
+
+class PromptRequest(BaseModel):
+    prompt: str = Field(
+        description="The original prompt to optimize"
+    )
+    goal: str = Field(
+        description="What the prompt should accomplish"
+    )
+    model_config = {"extra": "forbid"}
+
+class PromptResponse(BaseModel):
+    original_prompt: str = Field(
+        description="The original prompt that was submitted"
+    )
+    optimized_prompt: str = Field(
+        description="The improved version of the prompt"
+    )
+    changes: str = Field(
+        description="Explanation of what was improved and why"
+    )
+
+@app.post("/optimize", response_model=PromptResponse)
+def optimize_prompt_endpoint(request: PromptRequest):
+    try:
+        result = optimize_prompt(request.prompt, request.goal)
+    except ValueError as e:
+        logger.error(f"Optimization failed: {e}")
+        raise HTTPException(status_code=502, detail="Invalid upstream LLM response. Please retry.")
+    except Exception as e:
+        logger.error(f"Unexpected error: {e}")
+        raise HTTPException(
+            status_code=500,
+            detail="Internal server error. Please try again."
+        )
+
+    try:
+        log_optimization(
+            result["original_prompt"],
+            result["optimized_prompt"],
+            result["changes"],
+        )
+    except Exception as e:
+        logger.error(f"Failed to log optimization: {e}")
+
+    return PromptResponse(**result)

docker-compose-ai-eng/solution-files/requirements.txt

@@ -0,0 +1,6 @@
+fastapi==0.133.0
+uvicorn==0.40.0
+openai==2.26.0
+pydantic==2.11.7
+python-dotenv==1.1.0
+psycopg2-binary==2.9.10

docker-compose-ai-eng/solution-files/service.py

@@ -0,0 +1,60 @@
+from openai import OpenAI
+from dotenv import load_dotenv
+import json
+import os
+
+load_dotenv()
+
+client = OpenAI(
+    api_key=os.environ.get("OPENAI_API_KEY"),
+    base_url="https://api.openai.com/v1"
+)
+
+def optimize_prompt(prompt: str, goal: str) -> dict:
+    """Send a prompt to the LLM for optimization and return structured results."""
+
+    system_message = """You are a prompt engineering expert. Your job is to improve
+prompts so they produce better results from language models.
+
+You will receive an original prompt and a goal describing what the prompt should accomplish.
+Return your response as a JSON object with exactly these fields:
+- "optimized_prompt": the improved version of the prompt
+- "changes": a brief explanation of what you improved and why
+
+Return ONLY the JSON object. No markdown formatting, no extra text."""
+
+    user_message = f"""Original prompt: {prompt}
+
+Goal: {goal}
+
+Optimize this prompt to better achieve the stated goal."""
+
+    response = client.chat.completions.create(
+        model="gpt-4o-mini",
+        messages=[
+            {"role": "system", "content": system_message},
+            {"role": "user", "content": user_message}
+        ],
+        response_format={"type": "json_object"},
+        temperature=0.7
+    )
+
+    result_text = response.choices[0].message.content
+
+    try:
+        result = json.loads(result_text)
+    except json.JSONDecodeError:
+        raise ValueError(
+            f"LLM returned invalid JSON: {result_text[:200]}"
+        )
+
+    if "optimized_prompt" not in result or "changes" not in result:
+        raise ValueError(
+            f"LLM response missing required fields. Got: {list(result.keys())}"
+        )
+
+    return {
+        "original_prompt": prompt,
+        "optimized_prompt": result["optimized_prompt"],
+        "changes": result["changes"]
+    }

docker-compose-ai-eng/starter-files/Dockerfile

@@ -0,0 +1,10 @@
+FROM python:3.10-slim
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY main.py service.py ./
+
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

docker-compose-ai-eng/starter-files/main.py

@@ -0,0 +1,48 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel, Field
+from service import optimize_prompt
+import logging
+
+logger = logging.getLogger(__name__)
+
+app = FastAPI()
+
+@app.get("/health")
+def health_check():
+    return {"status": "ok"}
+
+class PromptRequest(BaseModel):
+    prompt: str = Field(
+        description="The original prompt to optimize"
+    )
+    goal: str = Field(
+        description="What the prompt should accomplish"
+    )
+    model_config = {"extra": "forbid"}
+
+class PromptResponse(BaseModel):
+    original_prompt: str = Field(
+        description="The original prompt that was submitted"
+    )
+    optimized_prompt: str = Field(
+        description="The improved version of the prompt"
+    )
+    changes: str = Field(
+        description="Explanation of what was improved and why"
+    )
+
+@app.post("/optimize", response_model=PromptResponse)
+def optimize_prompt_endpoint(request: PromptRequest):
+    try:
+        result = optimize_prompt(request.prompt, request.goal)
+    except ValueError as e:
+        logger.error(f"Optimization failed: {e}")
+        raise HTTPException(status_code=502, detail="Invalid upstream LLM response. Please retry.")
+    except Exception as e:
+        logger.error(f"Unexpected error: {e}")
+        raise HTTPException(
+            status_code=500,
+            detail="Internal server error. Please try again."
+        )
+
+    return PromptResponse(**result)

docker-compose-ai-eng/starter-files/requirements.txt

@@ -0,0 +1,5 @@
+fastapi==0.133.0
+uvicorn==0.40.0
+openai==2.26.0
+pydantic==2.11.7
+python-dotenv==1.1.0

docker-compose-ai-eng/starter-files/service.py

@@ -0,0 +1,60 @@
+from openai import OpenAI
+from dotenv import load_dotenv
+import json
+import os
+
+load_dotenv()
+
+client = OpenAI(
+    api_key=os.environ.get("OPENAI_API_KEY"),
+    base_url="https://api.openai.com/v1"
+)
+
+def optimize_prompt(prompt: str, goal: str) -> dict:
+    """Send a prompt to the LLM for optimization and return structured results."""
+
+    system_message = """You are a prompt engineering expert. Your job is to improve
+prompts so they produce better results from language models.
+
+You will receive an original prompt and a goal describing what the prompt should accomplish.
+Return your response as a JSON object with exactly these fields:
+- "optimized_prompt": the improved version of the prompt
+- "changes": a brief explanation of what you improved and why
+
+Return ONLY the JSON object. No markdown formatting, no extra text."""
+
+    user_message = f"""Original prompt: {prompt}
+
+Goal: {goal}
+
+Optimize this prompt to better achieve the stated goal."""
+
+    response = client.chat.completions.create(
+        model="gpt-4o-mini",
+        messages=[
+            {"role": "system", "content": system_message},
+            {"role": "user", "content": user_message}
+        ],
+        response_format={"type": "json_object"},
+        temperature=0.7
+    )
+
+    result_text = response.choices[0].message.content
+
+    try:
+        result = json.loads(result_text)
+    except json.JSONDecodeError:
+        raise ValueError(
+            f"LLM returned invalid JSON: {result_text[:200]}"
+        )
+
+    if "optimized_prompt" not in result or "changes" not in result:
+        raise ValueError(
+            f"LLM response missing required fields. Got: {list(result.keys())}"
+        )
+
+    return {
+        "original_prompt": prompt,
+        "optimized_prompt": result["optimized_prompt"],
+        "changes": result["changes"]
+    }