Build 100% Local Planning Agent with Qwen and LangGraph | Private Financial AI Agent with Ollama

2026-03-16 · Source: Venelin Valkov · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Intermediate, long

Summary

This content details the construction of a local-only planning agent designed for financial applications, utilizing a LangGraph architecture with four distinct nodes: a planner, an executor, a replanner, and a final reporter. The agent leverages custom tools for tasks like retrieving stock metrics, searching news, and comparing financial data, all managed through a tool registry. It operates on financial data, including stock metrics and hypothetical news headlines, and uses the Quantry 8 billion parameter model served via OAMA, with a context window of 16k tokens. The agent processes user queries, generates step-by-step plans, executes tool calls, and iteratively refines its plan before producing a final report, demonstrating its capabilities with queries on technology and healthcare stocks.

Key takeaway

For AI Engineers building financial analysis tools, consider adopting a LangGraph-based planning agent architecture. This approach allows for robust, iterative plan execution and reporting using local LLMs like Quantry 8B and custom data tools, enabling more structured and reliable financial insights. You should explore defining distinct agent roles (planner, executor, replanner, reporter) to manage complex queries effectively.

Key insights

A local planning agent uses LangGraph and custom tools for financial data analysis and report generation.

Principles

• Modular agent design improves task decomposition.
• Structured output enhances agent reliability.

Method

The agent workflow involves planning, executing custom tools, replanning based on execution results, and generating a final report, all orchestrated via LangGraph nodes and a tool registry.

In practice

• Implement custom tools for specific data access.
• Use LangGraph for complex agent orchestration.

Topics

• Planning Agents
• LangGraph
• Financial Analysis
• Custom Tools
• OAMA

Best for: AI Engineer, Machine Learning Engineer, AI Student

Related on AIssential

• Agentic AI in Action — Part 12 — Building a Deal Desk Intelligence Agent with LangChain and OpenAI — Combine LLMs for language interpretation with deterministic code for policy enforcement to build reliable enterprise AI agents.
• Build 100% Local Advanced RAG System for Financial PDFs with Qwen 3.5 | Docling, LangGraph & Ollama — A local RAG system can analyze financial PDFs using open-source tools for privacy and offline capability.
• Building an Agentic AI App With a Local LLM — No Cloud, No API Costs — Local LLMs on consumer GPUs enable cost-free, private AI app development with cloud-ready architecture.
• Your First Real LangGraph Project: — LangGraph enables building complex, stateful AI agents with memory, tools, and human oversight.
• Why No One Talks About LangChain, LangGraph Or AutoAgent — The post-framework AI architecture bifurcated into Pipeline-shaped for bounded workflows and Harness-shaped for open-ended agentic tasks.
• langchain-ai / open-swe — Open SWE provides an open-source framework for building internal coding agents, replicating architectures of leading tech companies.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Venelin Valkov.