ai-engineer

AI 工程

构建用于生产环境的大模型应用和 AI 系统。

适用场景

• 集成大模型 API
• 构建 RAG 系统
• 创建 AI Agent
• 配置向量数据库
• 优化 token 使用

大模型集成

API 配置

from anthropic import Anthropic

client = Anthropic()

def chat(messages: list[dict], system: str = None) -> str:
    response = client.messages.create(
        model="claude-sonnet-4-20250514",
        max_tokens=1024,
        system=system or "You are a helpful assistant.",
        messages=messages
    )
    return response.content[0].text

# With retry and error handling
from tenacity import retry, stop_after_attempt, wait_exponential

@retry(stop=stop_after_attempt(3), wait=wait_exponential(min=1, max=10))
def safe_chat(messages, system=None):
    try:
        return chat(messages, system)
    except Exception as e:
        logger.error(f"LLM call failed: {e}")
        raise

结构化输出

import json

def extract_structured(text: str, schema: dict) -> dict:
    prompt = f"""Extract information from the text according to this schema:
{json.dumps(schema, indent=2)}

Text: {text}

Return valid JSON only."""

    response = chat([{"role": "user", "content": prompt}])
    return json.loads(response)

RAG 系统

文档处理

from langchain.text_splitter import RecursiveCharacterTextSplitter

def chunk_documents(docs: list[str], chunk_size=1000, overlap=200):
    splitter = RecursiveCharacterTextSplitter(
        chunk_size=chunk_size,
        chunk_overlap=overlap,
        separators=["\n\n", "\n", ". ", " "]
    )
    return splitter.split_documents(docs)

向量存储

from qdrant_client import QdrantClient
from qdrant_client.models import Distance, VectorParams

client = QdrantClient(":memory:")  # or url="http://localhost:6333"

# Create collection
client.create_collection(
    collection_name="docs",
    vectors_config=VectorParams(size=1536, distance=Distance.COSINE)
)

# Upsert vectors
client.upsert(
    collection_name="docs",
    points=[
        {"id": i, "vector": embed(chunk), "payload": {"text": chunk}}
        for i, chunk in enumerate(chunks)
    ]
)

# Search
results = client.search(
    collection_name="docs",
    query_vector=embed(query),
    limit=5
)

RAG 查询

def rag_query(question: str, top_k=5) -> str:
    # Retrieve relevant chunks
    results = client.search(
        collection_name="docs",
        query_vector=embed(question),
        limit=top_k
    )

    context = "\n\n".join([r.payload["text"] for r in results])

    prompt = f"""Answer based on the context below.

Context:
{context}

Question: {question}

Answer:"""

    return chat([{"role": "user", "content": prompt}])

成本优化

• 缓存常见查询
• 对简单任务使用较小的模型
• 尽可能批量发送请求
• 按功能跟踪 token 使用量
• 合理设置 max_tokens

示例

输入：“为此应用添加 AI 聊天” **操作：**配置大模型客户端，创建聊天端点，添加错误处理

输入：“为文档构建 RAG” **操作：**对文档进行分块，创建嵌入，配置向量存储，实现搜索