Construct an AI Journal with LlamaIndex

will share the way to construct an AI journal with the LlamaIndex. We’ll cowl one important operate of this AI journal: asking for recommendation. We’ll begin with probably the most primary implementation and iterate from there. We will see important enhancements for this operate once we apply design patterns like Agentic Rag and multi-agent workflow.

You could find the supply code of this AI Journal in my GitHub repo right here. And about who I’m.

Overview of AI Journal

I wish to construct my rules by following Ray Dalio’s apply. An AI journal will assist me to self-reflect, observe my enchancment, and even give me recommendation. The general operate of such an AI journal appears to be like like this:

In the present day, we are going to solely cowl the implementation of the seek-advise stream, which is represented by a number of purple cycles within the above diagram.

Easiest Type: LLM with Giant Context

In probably the most easy implementation, we will cross all of the related content material into the context and fasten the query we wish to ask. We will try this in Llamaindex with a number of traces of code.

import pymupdf
from llama_index.llms.openai import OpenAI

path_to_pdf_book = './path/to/pdf/guide.pdf'
def load_book_content():
    textual content = ""
    with pymupdf.open(path_to_pdf_book) as pdf:
        for web page in pdf:
            textual content += str(web page.get_text().encode("utf8", errors='ignore'))
    return textual content

system_prompt_template = """You might be an AI assistant that gives considerate, sensible, and *deeply customized* options by combining:
- The person's private profile and rules
- Insights retrieved from *Ideas* by Ray Dalio
Guide Content material: 
```
{book_content}
```
Person profile:
```
{user_profile}
```
Person's query:
```
{user_question}
```
"""

def get_system_prompt(book_content: str, user_profile: str, user_question: str):
    system_prompt = system_prompt_template.format(
        book_content=book_content,
        user_profile=user_profile,
        user_question=user_question
    )
    return system_prompt

def chat():
    llm = get_openai_llm()
    user_profile = enter(">>Inform me about your self: ")
    user_question = enter(">>What do you wish to ask: ")
    user_profile = user_profile.strip()
    book_content = load_book_summary()
    response = llm.full(immediate=get_system_prompt(book_content, user_profile, user_question))
    return response

This method has downsides:

• Low Precision: Loading all of the guide context may immediate LLM to lose deal with the person’s query.
• Excessive Value: Sending over significant-sized content material in each LLM name means excessive value and poor efficiency.

With this method, if you happen to cross the entire content material of Ray Dalio’s Ideas guide, responses to questions like “ deal with stress?” develop into very common. Such responses with out referring to my query made me really feel that the AI was not listening to me. Despite the fact that it covers many vital ideas like embracing actuality, the 5-step course of to get what you need, and being radically open-minded. I like the recommendation I received to be extra focused to the query I raised. Let’s see how we will enhance it with RAG.

Enhanced Type: Agentic RAG

So, what’s Agentic RAG? Agentic RAG is combining dynamic decision-making and information retrieval. In our AI journal, the Agentic RAG stream appears to be like like this:

• Query Analysis: Poorly framed questions result in poor question outcomes. The agent will consider the person’s question and make clear the questions if the Agent believes it’s mandatory.
• Query Re-write: Rewrite the person enquiry to venture it to the listed content material within the semantic area. I discovered these steps important for bettering the precision in the course of the retrieval. Let’s say in case your information base is Q/A pair and you’re indexing the questions half to seek for solutions. Rewriting the person’s question assertion to a correct query will enable you discover probably the most related content material.
• Question Vector Index: Many parameters will be tuned when constructing such an index, together with chunk measurement, overlap, or a unique index kind. For simplicity, we’re utilizing VectorStoreIndex right here, which has a default chunking technique.
• Filter & Artificial: As a substitute of a fancy re-ranking course of, I explicitly instruct LLM to filter and discover related content material within the immediate. I see LLM choosing up probably the most related content material, although generally it has a decrease similarity rating than others.

With this Agentic RAG, you may retrieve extremely related content material to the person’s questions, producing extra focused recommendation.

Let’s study the implementation. With the LlamaIndex SDK, creating and persisting an index in your native listing is easy.

from llama_index.core import Doc, VectorStoreIndex, StorageContext, load_index_from_storage

Settings.embed_model = OpenAIEmbedding(api_key="ak-xxxx")
PERSISTED_INDEX_PATH = "/path/to/the/listing/persist/index/regionally"

def create_index(content material: str):
    paperwork = [Document(text=content)]
    vector_index = VectorStoreIndex.from_documents(paperwork)
    vector_index.storage_context.persist(persist_dir=PERSISTED_INDEX_PATH)

def load_index():
    storage_context = StorageContext.from_defaults(persist_dir=PERSISTED_INDEX_PATH)
    index = load_index_from_storage(storage_context)
    return index

As soon as we have now an index, we will create a question engine on prime of that. The question engine is a robust abstraction that permits you to alter the parameters in the course of the question(e.g., TOP Okay) and the synthesis behaviour after the content material retrieval. In my implementation, I overwrite the response_mode NO_TEXT as a result of the agent will course of the guide content material returned by the operate name and synthesize the ultimate consequence. Having the question engine to synthesize the consequence earlier than passing it to the agent could be redundant.

from llama_index.core.indices.vector_store import VectorIndexRetriever
from llama_index.core.query_engine import RetrieverQueryEngine
from llama_index.core.response_synthesizers import ResponseMode
from llama_index.core import  VectorStoreIndex, get_response_synthesizer

def _create_query_engine_from_index(index: VectorStoreIndex):
    # configure retriever
    retriever = VectorIndexRetriever(
        index=index,
        similarity_top_k=TOP_K,
    )
    # return the unique content material with out utilizing LLM to synthesizer. For later analysis.
    response_synthesizer = get_response_synthesizer(response_mode=ResponseMode.NO_TEXT)
    # assemble question engine
    query_engine = RetrieverQueryEngine(
        retriever=retriever,
        response_synthesizer=response_synthesizer
    )
    return query_engine

The immediate appears to be like like the next:

You might be an assistant that helps reframe person questions into clear, concept-driven statements that match 
the fashion and matters of Ideas by Ray Dalio, and carry out search for precept guide for related content material. 

Background:
Ideas teaches structured enthusiastic about life and work selections.
The important thing concepts are:
* Radical reality and radical transparency
* Determination-making frameworks
* Embracing errors as studying

Process:
- Process 1: Make clear the person's query if wanted. Ask follow-up questions to make sure you perceive the person's intent.
- Process 2: Rewrite a person’s query into a press release that might match how Ray Dalio frames concepts in Ideas. Use formal, logical, impartial tone.
- Process 3: Lookup precept guide with given re-wrote statements. You must present at the very least {REWRITE_FACTOR} rewrote variations.
- Process 4: Discover probably the most related from the guide content material as your fina solutions.

Lastly, we will construct the agent with these capabilities outlined.

def get_principle_rag_agent():
    index = load_persisted_index()
    query_engine = _create_query_engine_from_index(index)

    def look_up_principle_book(original_question: str, rewrote_statement: Record[str]) -> Record[str]:
        consequence = []
        for q in rewrote_statement:
            response = query_engine.question(q)
            content material = [n.get_content() for n in response.source_nodes]
            consequence.prolong(content material)
        return consequence

    def clarify_question(original_question: str, your_questions_to_user: Record[str]) -> str:
        """
        Make clear the person's query if wanted. Ask follow-up questions to make sure you perceive the person's intent.
        """
        response = ""
        for q in your_questions_to_user:
            print(f"Query: {q}")
            r = enter("Response:")
            response += f"Query: {q}nResponse: {r}n"
        return response

    instruments = [
        FunctionTool.from_defaults(
            fn=look_up_principle_book,
            name="look_up_principle_book",
            description="Look up principle book with re-wrote queries. Getting the suggestions from the Principle book by Ray Dalio"),
        FunctionTool.from_defaults(
            fn=clarify_question,
            name="clarify_question",
            description="Clarify the user's question if needed. Ask follow-up questions to ensure you understand the user's intent.",
        )
    ]

    agent = FunctionAgent(
        identify="principle_reference_loader",
        description="You're a useful agent will based mostly on person's query and search for probably the most related content material in precept guide.n",
        system_prompt=QUESTION_REWRITE_PROMPT,
        instruments=instruments,
    )
    return agent

rag_agent = get_principle_rag_agent()
response = await agent.run(chat_history=chat_history)

There are a number of observations I had in the course of the implementations:

• One fascinating reality I discovered is that offering a non-used parameter, original_question , within the operate signature helps. I discovered that after I shouldn’t have such a parameter, LLM generally doesn’t observe the rewrite instruction and passes the unique query in rewrote_statement the parameter. Having original_question parameters by some means emphasizes the rewriting mission to LLM.
• Completely different LLMs behave fairly otherwise given the identical immediate. I discovered DeepSeek V3 far more reluctant to set off operate calls than different mannequin suppliers. This doesn’t essentially imply it isn’t usable. If a practical name must be initiated 90% of the time, it must be a part of the workflow as an alternative of being registered as a operate name. Additionally, in comparison with OpenAI’s fashions, I discovered Gemini good at citing the supply of the guide when it synthesizes the outcomes.
• The extra content material you load into the context window, the extra inference functionality the mannequin wants. A smaller mannequin with much less inference energy is extra prone to get misplaced within the giant context offered.

Nonetheless, to finish the seek-advice operate, you’ll want a number of Brokers working collectively as an alternative of a single Agent. Let’s speak about the way to chain your Brokers collectively into workflows.

Closing Type: Agent Workflow

Earlier than we begin, I like to recommend this text by Anthropic, Constructing Efficient Brokers. The one-liner abstract of the articles is that you must at all times prioritise constructing a workflow as an alternative of a dynamic agent when potential. In LlamaIndex, you are able to do each. It permits you to create an agent workflow with extra computerized routing or a customized workflow with extra specific management of the transition of steps. I’ll present an instance of each implementations.

Let’s check out how one can construct a dynamic workflow. Here’s a code instance.

interviewer = FunctionAgent(
        identify="interviewer",
        description="Helpful agent to make clear person's questions",
        system_prompt=_intervierw_prompt,
        can_handoff_to = ["retriver"]
        instruments=instruments
)
interviewer = FunctionAgent(
        identify="retriever",
        description="Helpful agent to retrive precept guide's content material.",
        system_prompt=_retriver_prompt,
        can_handoff_to = ["advisor"]
        instruments=instruments
)
advisor = FunctionAgent(
        identify="advisor",
        description="Helpful agent to advise person.",
        system_prompt=_advisor_prompt,
        can_handoff_to = []
        instruments=instruments
)
workflow = AgentWorkflow(
        brokers=[interviewer, advisor, retriever],
        root_agent="interviewer",
    )
handler = await workflow.run(user_msg=" deal with stress?")

It’s dynamic as a result of the Agent transition is predicated on the operate name of the LLM mannequin. Underlying, LlamaIndex workflow gives agent descriptions as capabilities for LLM fashions. When the LLM mannequin triggers such “Agent Operate Name”, LlamaIndex will path to your subsequent corresponding agent for the following step processing. Your earlier agent’s output has been added to the workflow inside state, and your following agent will choose up the state as a part of the context of their name to the LLM mannequin. You additionally leverage state and reminiscence parts to handle the workflow’s inside state or load exterior information(reference the doc right here).

Nonetheless, as I’ve urged, you may explicitly management the steps in your workflow to achieve extra management. With LlamaIndex, it may be performed by extending the workflow object. For instance:

class ReferenceRetrivalEvent(Occasion):
    query: str

class Recommendation(Occasion):
    rules: Record[str]
    profile: dict
    query: str
    book_content: str

class AdviceWorkFlow(Workflow):
    def __init__(self, verbose: bool = False, session_id: str = None):
        state = get_workflow_state(session_id)
        self.rules = state.load_principle_from_cases()
        self.profile = state.load_profile()
        self.verbose = verbose
        tremendous().__init__(timeout=None, verbose=verbose)

    @step
    async def interview(self, ctx: Context,
                        ev: StartEvent) -> ReferenceRetrivalEvent:
        # Step 1: Interviewer agent asks inquiries to the person
        interviewer = get_interviewer_agent()
        query = await _run_agent(interviewer, query=ev.user_msg, verbose=self.verbose)

        return ReferenceRetrivalEvent(query=query)

    @step
    async def retrieve(self, ctx: Context, ev: ReferenceRetrivalEvent) -> Recommendation:
        # Step 2: RAG agent retrieves related content material from the guide
        rag_agent = get_principle_rag_agent()
        book_content = await _run_agent(rag_agent, query=ev.query, verbose=self.verbose)
        return Recommendation(rules=self.rules, profile=self.profile,
                      query=ev.query, book_content=book_content)

    @step
    async def recommendation(self, ctx: Context, ev: Recommendation) -> StopEvent:
        # Step 3: Adviser agent gives recommendation based mostly on the person's profile, rules, and guide content material
        advisor = get_adviser_agent(ev.profile, ev.rules, ev.book_content)
        advise = await _run_agent(advisor, query=ev.query, verbose=self.verbose)
        return StopEvent(consequence=advise)

The particular occasion kind’s return controls the workflow’s step transition. As an example, retrieve step returns an Recommendation occasion that may set off the execution of the recommendation step. You too can leverage the Recommendation occasion to cross the required info you want.

Throughout the implementation, if you’re irritated by having to begin over the workflow to debug some steps within the center, the context object is crucial once you wish to failover the workflow execution. You’ll be able to retailer your state in a serialised format and get well your workflow by unserialising it to a context object. Your workflow will proceed executing based mostly on the state as an alternative of beginning over.

workflow = AgentWorkflow(
    brokers=[interviewer, advisor, retriever],
    root_agent="interviewer",
)
attempt:
    handler = w.run()
    consequence = await handler
besides Exception as e:
    print(f"Error throughout preliminary run: {e}")
    await fail_over()
    # Non-obligatory, serialised and save the contexct for debugging 
    ctx_dict = ctx.to_dict(serializer=JsonSerializer())
    json_dump_and_save(ctx_dict)
    # Resume from the identical context
    ctx_dict = load_failed_dict()
    restored_ctx = Context.from_dict(workflow, ctx_dict,serializer=JsonSerializer())
    handler = w.run(ctx=handler.ctx)
    consequence = await handler

Abstract

On this put up, we have now mentioned the way to use LlamaIndex to implement an AI journal’s core operate. The important thing studying consists of:

• Utilizing Agentic RAG to leverage LLM functionality to dynamically rewrite the unique question and synthesis consequence.
• Use a Custom-made Workflow to achieve extra specific management over step transitions. Construct dynamic brokers when mandatory.

The bitterce code of this AI journal is in my GitHub repo right here. I hope you take pleasure in this text and this small app I constructed. Cheers!