Assignment Requirement Document | Build an AI Financial Analyst Using LangGraph | Major Project

Course

Advanced Artificial Intelligence Systems EngineeringM.Tech / MCA / B.Tech (AI & ML Specialization) / Advanced Full Stack AI Development

Assignment Category

Capstone Assignment / Major Semester Project

Assignment Duration

6–8 Weeks

Difficulty Level

Advanced to Industry-Level

Introduction

Financial institutions, investment firms, hedge funds, and fintech startups are increasingly adopting Large Language Models (LLMs), autonomous agents, and AI orchestration systems to automate financial analysis and decision-making workflows.

In this assignment, students are required to design and develop a fully functional AI-powered Financial Analyst platform using LangGraph and modern AI engineering practices.

The project should simulate a real-world financial analysis assistant capable of:

• Performing stock market analysis

• Computing technical indicators

• Analyzing financial trends

• Summarizing financial news

• Evaluating portfolios

• Performing sentiment analysis

• Generating AI-driven financial reports

• Conducting multi-step reasoning workflows

• Managing autonomous agent collaboration

This assignment is intentionally designed to resemble real-world fintech AI systems and production-grade agentic architectures.

Students are expected to demonstrate:

• Software engineering skills

• AI orchestration capabilities

• Financial analytics understanding

• API integrations

• Data visualization

• Multi-agent system design

• Scalable architecture practices

Assignment Objectives

The purpose of this assignment is to help students gain practical experience in:

• Building AI agent workflows

• Using LangGraph for orchestration

• Implementing stateful AI systems

• Integrating real-time financial APIs

• Designing modular AI architectures

• Creating production-style fintech applications

• Building autonomous financial reasoning systems

• Working with structured and unstructured financial data

• Applying AI to financial intelligence workflows

Project Scenario

You are working as an AI Engineer for a fintech startup building an AI Financial Intelligence Platform for retail investors.

The startup wants an AI system that can:

• Analyze stock performance

• Explain market behavior

• Generate portfolio recommendations

• Track financial sentiment

• Interpret technical indicators

• Produce AI-generated investment insights

Your task is to design and implement this system using LangGraph-based agent orchestration.

Core Project Requirements

Students must implement ALL mandatory modules listed below.

MODULE 1 — User Interaction & Query

Understanding

Objective

Create a system capable of understanding natural language financial queries.

Functional Requirements

The system should support queries such as:

• “Analyze Tesla stock for the past 6 months”

• “Compare Apple and Microsoft performance”

• “Summarize the latest NVIDIA news”

• “Evaluate my portfolio risk”

• “Show RSI and MACD trends for Amazon”

• “Which stocks are currently overbought?”

Required Features

Query Classification

The system should identify:

• Stock analysis requests

• Portfolio analysis requests

• News-related requests

• Comparative analysis requests

• Technical indicator requests

Intent Routing

The application should route queries to appropriate LangGraph agents.

Conversation Memory

Implement memory/state handling for:

• Previous stock searches

• User portfolio history

• Historical analysis requests

• Follow-up questions

Expected Deliverables

• NLP query parser

• Intent classification module

• Stateful memory workflow

• Query routing logic

MODULE 2 — Financial Data Collection Engine

Objective

Build a robust financial data ingestion system.

Required Features

Stock Data Retrieval

Fetch:

• Historical prices

• OHLC data

• Intraday data

• Trading volume

• Dividend information

• Market capitalization

• PE ratio

• Beta values

Multi-Stock Comparison

Support:

• Comparative charts

• Relative performance analysis

• Correlation analysis

Market Data Handling

The system should support:

• NASDAQ stocks

• NYSE stocks

• ETFs

• Crypto assets (optional)

Suggested APIs

Students may use:

• Alpha Vantage

• Yahoo Finance API (yfinance)

• Finnhub

• Polygon.io

Additional Engineering Requirements

Students must implement:

• API retry mechanisms

• Rate-limit handling

• Caching

• Error recovery

• Missing data handling

MODULE 3 — Technical Analysis Engine

Objective

Develop an advanced technical analysis system.

Mandatory Technical Indicators

Students MUST implement ALL the following:

Indicator	Required
SMA	Yes
EMA	Yes
RSI	Yes
MACD	Yes
Bollinger Bands	Yes
ATR	Yes
VWAP	Yes
OBV	Yes

Additional Optional Indicators

Students may additionally implement:

• Ichimoku Cloud

• Fibonacci Retracement

• SuperTrend

• ADX

• Parabolic SAR

• Momentum Oscillator

Functional Requirements

For each indicator, the system should:

• Compute values correctly

• Generate charts

• Explain indicator meaning

• Detect bullish/bearish signals

• Detect overbought/oversold conditions

• Produce AI-generated interpretations

Required AI Capabilities

The AI system should explain:

• Why a signal is bullish/bearish

• Trend reversals

• Momentum changes

• Volatility behavior

• Risk indications

Visualization Requirements

Students must implement:

• Candlestick charts

• Indicator overlays

• Zoomable graphs

• Comparative charts

• Multi-timeframe visualizations

MODULE 4 — Financial News

Intelligence System

Objective

Build an AI-powered financial news analysis engine.

Functional Requirements

The system should:

• Fetch financial news articles

• Summarize long-form articles

• Perform sentiment analysis

• Detect stock-specific sentiment

• Identify major market events

• Detect trending topics

Required Features

News Aggregation

Collect news from multiple sources.

Summarization

Generate:

• Short summaries

• Detailed summaries

• Bullet-point insights

Sentiment Analysis

Classify:

• Positive sentiment

• Negative sentiment

• Neutral sentiment

The model should also provide:

• Confidence score

• Reasoning behind sentiment

Duplicate Detection

Students must remove duplicate or near-duplicate news articles.

News-to-Stock Mapping

Associate articles with:

• Stock tickers

• Companies

• Market sectors

Suggested APIs

• NewsAPI

• GNews

• Finnhub News APIs

MODULE 5 — Portfolio Intelligence System

Objective

Develop a portfolio analysis and insight generation engine.

Required Features

The system should:

• Accept portfolio inputs

• Analyze asset allocation

• Detect concentration risk

• Compute diversification scores

• Analyze sector exposure

• Detect volatility exposure

• Generate AI investment observations

Example Portfolio Input

[
  {
    "ticker": "AAPL",
    "quantity": 15
  },
  {
    "ticker": "NVDA",
    "quantity": 8
  },
  {
    "ticker": "MSFT",
    "quantity": 10
  }
]

AI Insight Requirements

The AI should generate:

• Portfolio strengths

• Portfolio weaknesses

• Diversification recommendations

• Sector imbalance warnings

• Risk exposure commentary

• Growth vs value observations

• Volatility warnings

Advanced Requirements

Students should additionally implement:

• Portfolio performance tracking

• Historical portfolio growth

• Benchmark comparison (S&P 500 etc.)

• Sharpe ratio (optional)

• Beta analysis (optional)

MODULE 6 — LangGraph Multi-Agent

Workflow

Objective

Design a production-style multi-agent orchestration workflow.

Mandatory LangGraph Agents

Students MUST implement the following agents:

Agent	Responsibility
Query Understanding Agent	Understand user requests
Financial Data Agent	Fetch stock/market data
Technical Analysis Agent	Compute indicators
News Intelligence Agent	Analyze financial news
Portfolio Analysis Agent	Evaluate portfolio
Report Generation Agent	Generate final report
Memory Agent	Maintain state/context
Validation Agent	Validate outputs

LangGraph Requirements

Students MUST implement:

• Stateful graph execution

• Node-based orchestration

• Conditional routing

• Retry logic

• Error handling

• Agent collaboration

• Structured outputs

• Tool invocation workflows

Advanced Workflow Features

Students are encouraged to implement:

• Recursive reasoning

• Reflection agents

• Self-correction workflows

• Multi-agent debate systems

• Confidence-based routing

MODULE 7 — AI Report Generation

Objective

Generate professional AI-driven financial reports.

Required Report Sections

The generated report should include:

• Executive summary

• Stock trend overview

• Technical analysis summary

• Market sentiment overview

• Portfolio observations

• Risk analysis

• AI recommendations

• Conclusion

Report Formats

Students must support:

• Markdown report generation

• PDF export

• Downloadable summaries

MODULE 8 — Frontend Dashboard

Objective

Build a professional fintech dashboard interface.

Mandatory UI Components

Students MUST implement:

• Login page

• Dashboard homepage

• Stock search page

• Portfolio dashboard

• News analysis panel

• AI insights section

• Technical indicator charts

• Historical trend charts

UI/UX Expectations

The interface should:

• Be responsive

• Use clean fintech-style design

• Include loading states

• Include error states

• Support dark/light themes

• Provide interactive visualizations

Suggested Frontend Stack

Students may use:

• React

• Next.js

• Streamlit

• Tailwind CSS

• Plotly

• Recharts

MODULE 9 — Database & Persistence Layer

Objective

Implement persistent storage and session handling.

Required Features

Students must store:

• User portfolios

• Historical queries

• Cached stock data

• AI-generated reports

• News summaries

Suggested Databases

• PostgreSQL

• MongoDB

• SQLite

• Redis (optional caching)

MODULE 10 — Deployment & DevOps

Objective

Deploy the system in a production-style environment.

Required Features

Students must:

• Containerize the application

• Use environment variables

• Implement logging

• Configure API secrets securely

• Provide deployment instructions

Suggested Deployment Platforms

• Render

• Railway

• Vercel

• AWS

Mandatory Technical Requirements

Students MUST demonstrate:

• Modular architecture

• Separation of concerns

• Reusable services

• API abstraction layers

• Type-safe schemas

• Environment configuration

• Proper documentation

• Exception handling

• Logging mechanisms

Expected Folder Structure

/ai-financial-analyst/
│
├── backend/
│   ├── agents/
│   ├── graphs/
│   ├── services/
│   ├── indicators/
│   ├── api/
│   ├── models/
│   ├── utils/
│   └── memory/
│
├── frontend/
│   ├── components/
│   ├── pages/
│   ├── charts/
│   ├── hooks/
│   └── services/
│
├── database/
├── docs/
├── reports/
├── tests/
├── docker/
├── README.md
├── requirements.txt
└── docker-compose.yml

Deliverables

Students must submit ALL of the following.

1. Complete Source Code

The repository should contain:

• Backend

• Frontend

• LangGraph workflows

• API integrations

• Documentation

• Configurations

2. Technical Report (15–25 Pages)

The report MUST include:

• Problem statement

• System architecture

• Workflow diagrams

• LangGraph explanation

• Technical analysis explanation

• AI pipeline design

• API integrations

• Testing methodologies

• Challenges faced

• Future scope

• Screenshots

3. Presentation Slides

Minimum:

• 20–30 slides

Must contain:

• Architecture diagrams

• Workflow diagrams

• Agent collaboration flow

• Feature demonstrations

• Technical stack

• Challenges

• Learnings

4. Demo Video

Duration:

• 10–15 minutes

The demo should showcase:

• End-to-end system flow

• LangGraph orchestration

• Technical indicator analysis

• News summarization

• Portfolio insights

• Dashboard functionality

5. Deployment Link

Students must provide:

• Live deployed application URL

• GitHub repository URL

Evaluation Rubric

Component	Weightage
LangGraph Architecture	15%
AI Agent Design	15%
Financial Analysis Accuracy	15%
Technical Indicators	10%
News Intelligence System	10%
Portfolio Analytics	10%
Frontend Dashboard	10%
Code Quality & Scalability	5%
Documentation & Report	5%
Deployment & DevOps	3%
Innovation & Advanced Features	2%

Bonus Features (Extra Credit)

Students may receive bonus marks for implementing:

• Autonomous investment advisor

• AI voice assistant

• Real-time websocket streaming

• RAG-based financial QA

• Vector database memory

• Multi-user support

• Role-based authentication

• AI recommendation engine

• Trading simulator

• AI-generated market alerts

• Options analysis

• Crypto analytics

• Agent self-reflection systems

• Multi-modal financial analysis

Coding Standards

Students are expected to:

• Follow clean coding principles

• Use meaningful variable names

• Write modular functions

• Avoid code duplication

• Use type annotations

• Add comments where necessary

• Follow consistent formatting

Prohibited Practices

Students MUST NOT:

• Hardcode credentials

• Use plagiarized repositories

• Fabricate financial results

• Submit incomplete workflows

• Ignore edge cases

• Use fake AI outputs

• Disable error handling for demo purposes

Academic Integrity Policy

This assignment is intended to evaluate:

• AI engineering skills

• System architecture understanding

• Problem-solving capabilities

• Financial analytics reasoning

Students may use:

• Open-source libraries

• Public APIs

• Official documentation

However, plagiarism from:

• GitHub repositories

• Online tutorials

• Commercial templates

• Other student submissions

  
  

will result in academic penalties.

Faculty members may conduct:

• Viva examinations

• Live coding verification

• Code walkthrough sessions

  
  

Students must be able to explain every implemented module.

Suggested Timeline

Week	Tasks
Week 1	Requirement analysis + architecture planning
Week 2	Financial data integration
Week 3	Technical indicators engine
Week 4	News intelligence system
Week 5	Portfolio analytics
Week 6	LangGraph orchestration
Week 7	Frontend dashboard + testing
Week 8	Deployment + documentation + final polishing

Recommended Learning Resources

• LangGraph Documentation

• LangChain Documentation

• OpenAI API Docs

• Pandas Documentation

• TA-Lib Documentation

• Plotly Documentation

Expected Learning Outcomes

Upon successful completion of this assignment, students should be capable of:

• Building production-style AI applications

• Designing multi-agent systems

• Using LangGraph professionally

• Integrating financial APIs

• Creating fintech dashboards

• Implementing AI reasoning pipelines

• Building scalable AI architectures

• Deploying full-stack AI systems

This project is intentionally designed to simulate real-world AI engineering challenges commonly encountered in:

• Fintech startups

• Quantitative trading firms

• AI research labs

• Financial analytics platforms

• Investment intelligence systems

• Autonomous AI product teams

---

Need guidance with this assignment or similar AI/ML projects?

CodersArts provides support for industry-level assignments, AI project development, LangGraph workflows, LLM applications, full-stack AI systems, fintech analytics projects, documentation, deployment, and technical mentorship.

Get help with:

• LangGraph & AI Agent Development

• Financial Analytics Projects

• LLM & Generative AI Applications

• Full-Stack AI Systems

• Technical Reports & Documentation

• Deployment & Architecture Guidance

• Advanced AI/ML Assignments