Vibe Coding a Personal AI Monetary Analyst with Python and Native LLMs

# Introduction

 
Final month, I discovered myself gazing my financial institution assertion, attempting to determine the place my cash was truly going. Spreadsheets felt cumbersome. Present apps are like black bins, and the worst half is that they demand I add my delicate monetary knowledge to a cloud server. I wished one thing completely different. I wished an AI knowledge analyst that might analyze my spending, spot uncommon transactions, and provides me clear insights — all whereas preserving my knowledge 100% native. So, I constructed one.

What began as a weekend challenge become a deep dive into real-world knowledge preprocessing, sensible machine studying, and the ability of native giant language fashions (LLMs). On this article, I’ll stroll you thru how I created an AI-powered monetary evaluation app utilizing Python with “Vibe Coding.” Alongside the way in which, you’ll be taught many sensible ideas that apply to any knowledge science challenge, whether or not you might be analyzing gross sales logs, sensor knowledge, or buyer suggestions.

By the top, you’ll perceive:

• How you can construct a strong knowledge preprocessing pipeline that handles messy, real-world CSV information
• How to decide on and implement machine studying fashions when you could have restricted coaching knowledge
• How you can design interactive visualizations that truly reply consumer questions
• How you can combine an area LLM for producing natural-language insights with out sacrificing privateness

The entire supply code is on the market on GitHub. Be at liberty to fork it, prolong it, or use it as a place to begin in your personal AI knowledge analyst.

Fig. 1: App dashboard displaying spending breakdown and AI insights | Picture by Writer

# The Downside: Why I Constructed This

 
Most private finance apps share a basic flaw: your knowledge leaves your management. You add financial institution statements to companies that retailer, course of, and doubtlessly monetize your info. I wished a device that:

1. Let me add and analyze knowledge immediately
2. Processed every little thing domestically — no cloud, no knowledge leaks
3. Supplied AI-powered insights, not simply static charts

This challenge grew to become my car for studying a number of ideas that each knowledge scientist ought to know, like dealing with inconsistent knowledge codecs, deciding on algorithms that work with small datasets, and constructing privacy-preserving AI options.

# Challenge Structure

 
Earlier than diving into code, here’s a challenge construction displaying how the items match collectively:

challenge/   
  ├── app.py              # Essential Streamlit app
  ├── config.py           # Settings (classes, Ollama config)
  ├── preprocessing.py    # Auto-detect CSV codecs, normalize knowledge
  ├── ml_models.py        # Transaction classifier + Isolation Forest anomaly detector
  ├── visualizations.py   # Plotly charts (pie, bar, timeline, heatmap)
  ├── llm_integration.py  # Ollama streaming integration
  ├── necessities.txt    # Dependencies
  ├── README.md           # Documentation with "deep dive" classes
  └── sample_data/
    ├── sample_bank_statement.csv
    └── sample_bank_format_2.csv

We’ll take a look at constructing every layer step-by-step.

# Step 1: Constructing a Sturdy Knowledge Preprocessing Pipeline

 
The primary lesson I discovered was that real-world knowledge is messy. Totally different banks export CSVs in utterly completely different codecs. Chase Financial institution makes use of “Transaction Date” and “Amount.” Financial institution of America makes use of “Date,” “Payee,” and separate “Debit” columns. Moniepoint and OPay every have their very own kinds.

A preprocessing pipeline should deal with these variations robotically.

// Auto-Detecting Column Mappings

I constructed a pattern-matching system that identifies columns no matter naming conventions. Utilizing common expressions, we are able to map unclear column names to plain fields.

import re

COLUMN_PATTERNS = {
    "date": [r"date", r"trans.*date", r"posting.*date"],
    "description": [r"description", r"memo", r"payee", r"merchant"],
    "amount": [r"^amount$", r"transaction.*amount"],
    "debit": [r"debit", r"withdrawal", r"expense"],
    "credit": [r"credit", r"deposit", r"income"],
}

def detect_column_mapping(df):
    mapping = {}
    for discipline, patterns in COLUMN_PATTERNS.objects():
        for col in df.columns:
            for sample in patterns:
                if re.search(sample, col.decrease()):
                    mapping[field] = col
                    break
    return mapping

The important thing perception: design for variations, not particular codecs. This strategy works for any CSV that makes use of widespread monetary phrases.

// Normalizing to a Commonplace Schema

As soon as columns are detected, we normalize every little thing right into a constant construction. For instance, banks that cut up debits and credit must be mixed right into a single quantity column (damaging for bills, optimistic for earnings):

if "debit" in mapping and "credit" in mapping:
    debit = df[mapping["debit"]].apply(parse_amount).abs() * -1
    credit score = df[mapping["credit"]].apply(parse_amount).abs()
    normalized["amount"] = credit score + debit

Key takeaway: Normalize your knowledge as quickly as potential. It simplifies each following operation, like characteristic engineering, machine studying modeling, and visualization.

Fig 2: The preprocessing report exhibits what the pipeline detected, giving customers transparency | Picture by Writer

# Step 2: Selecting Machine Studying Fashions for Restricted Knowledge

 
The second main problem is restricted coaching knowledge. Customers add their very own statements, and there’s no huge labeled dataset to coach a deep studying mannequin. We’d like algorithms that work effectively with small samples and will be augmented with easy guidelines.

// Transaction Classification: A Hybrid Strategy

As an alternative of pure machine studying, I constructed a hybrid system:

1. Rule-based matching for assured circumstances (e.g., key phrases like “WALMART” → groceries)
2. Sample-based fallback for ambiguous transactions

SPENDING_CATEGORIES = {
    "groceries": ["walmart", "costco", "whole foods", "kroger"],
    "dining": ["restaurant", "starbucks", "mcdonald", "doordash"],
    "transportation": ["uber", "lyft", "shell", "chevron", "gas"],
    # ... extra classes
}

def classify_transaction(description, quantity):
    for class, key phrases in SPENDING_CATEGORIES.objects():
        if any(kw in description.decrease() for kw in key phrases):
            return class
    return "income" if quantity > 0 else "other"

This strategy works instantly with none coaching knowledge, and it’s straightforward for customers to grasp and customise.

// Anomaly Detection: Why Isolation Forest?

For detecting uncommon spending, I wanted an algorithm that might:

1. Work with small datasets (not like deep studying)
2. Make no assumptions about knowledge distribution (not like statistical strategies like Z-score alone)
3. Present quick predictions for an interactive UI

Isolation Forest from scikit-learn ticked all of the bins. It isolates anomalies by randomly partitioning the info. Anomalies are few and completely different, in order that they require fewer splits to isolate.

from sklearn.ensemble import IsolationForest

detector = IsolationForest(
    contamination=0.05,  # Count on ~5% anomalies
    random_state=42
)
detector.match(options)
predictions = detector.predict(options)  # -1 = anomaly

I additionally mixed this with easy Z-score checks to catch apparent outliers. A Z-score describes the place of a uncooked rating by way of its distance from the imply, measured in normal deviations:
[
z = frac{x – mu}{sigma}
]
The mixed strategy catches extra anomalies than both technique alone.

Key takeaway: Generally easy, well-chosen algorithms outperform complicated ones, particularly when you could have restricted knowledge.

Fig 3: The anomaly detector flags uncommon transactions, which stand out within the timeline | Picture by Writer

# Step 3: Designing Visualizations That Reply Questions

 
Visualizations ought to reply questions, not simply present knowledge. I used Plotly for interactive charts as a result of it permits customers to discover the info themselves. Listed here are the design rules I adopted:

1. Constant colour coding: Purple for bills, inexperienced for earnings
2. Context by means of comparability: Present earnings vs. bills facet by facet
3. Progressive disclosure: Present a abstract first, then let customers drill down

For instance, the spending breakdown makes use of a donut chart with a gap within the center for a cleaner look:

import plotly.categorical as px

fig = px.pie(
    category_totals,
    values="Amount",
    names="Category",
    gap=0.4,
    color_discrete_map=CATEGORY_COLORS
)

Streamlit makes it straightforward so as to add these charts with st.plotly_chart() and construct a responsive dashboard.

Fig 4: A number of chart sorts give customers completely different views on the identical knowledge | Picture by Writer

# Step 4: Integrating a Native Giant Language Mannequin for Pure Language Insights

 
The ultimate piece was producing human-readable insights. I selected to combine Ollama, a device for operating LLMs domestically. Why native as an alternative of calling OpenAI or Claude?

1. Privateness: Financial institution knowledge by no means leaves the machine
2. Value: Limitless queries, zero API charges
3. Pace: No community latency (although technology nonetheless takes a couple of seconds)

// Streaming for Higher Person Expertise

LLMs can take a number of seconds to generate a response. Streamlit exhibits tokens as they arrive, making the wait really feel shorter. Right here is an easy implementation utilizing requests with streaming:

import requests
import json

def generate(self, immediate):
    response = requests.publish(
        f"{self.base_url}/api/generate",
        json={"model": "llama3.2", "prompt": immediate, "stream": True},
        stream=True
    )
    for line in response.iter_lines():
        if line:
            knowledge = json.hundreds(line)
            yield knowledge.get("response", "")

In Streamlit, you’ll be able to show this with st.write_stream().

st.write_stream(llm.get_overall_insights(df))

// Immediate Engineering for Monetary Knowledge

The important thing to helpful LLM output is a structured immediate that features precise knowledge. For instance:

immediate = f"""Analyze this monetary abstract:
- Complete Revenue: ${earnings:,.2f}
- Complete Bills: ${bills:,.2f}
- High Class: {top_category}
- Largest Anomaly: {anomaly_desc}

Present 2-3 actionable suggestions based mostly on this knowledge."""

This offers the mannequin concrete numbers to work with, resulting in extra related insights.

Fig 5: The add interface is straightforward; select a CSV and let the AI do the remaining | Picture by Writer

// Working the Software

Getting began is simple. You will want Python put in, then run:

pip set up -r necessities.txt

# Non-obligatory, for AI insights
ollama pull llama3.2

streamlit run app.py

Add any financial institution CSV (the app auto-detects the format), and inside seconds, you will notice a dashboard with categorized transactions, anomalies, and AI-generated insights.

# Conclusion

 
This challenge taught me that constructing one thing useful is just the start. The actual studying occurred after I requested why every bit works:

• Why auto-detect columns? As a result of real-world knowledge doesn’t comply with your schema. Constructing a versatile pipeline saves hours of guide cleanup.
• Why Isolation Forest? As a result of small datasets want algorithms designed for them. You don’t at all times want deep studying.
• Why native LLMs? As a result of privateness and price matter in manufacturing. Working fashions domestically is now sensible and highly effective.

These classes apply far past private finance, whether or not you might be analyzing gross sales knowledge, server logs, or scientific measurements. The identical rules of strong preprocessing, pragmatic modeling, and privacy-aware AI will serve you in any knowledge challenge.

The entire supply code is on the market on GitHub. Fork it, prolong it, and make it your individual. In case you construct one thing cool with it, I’d love to listen to about it.

// References

Shittu Olumide is a software program engineer and technical author keen about leveraging cutting-edge applied sciences to craft compelling narratives, with a eager eye for element and a knack for simplifying complicated ideas. You may also discover Shittu on Twitter.