• Go to Database AI & SQL Apps Landing Page
• Download PDF Version

This HTML page was converted from the original PDF. Some images and complex layouts may not appear correctly. For the best experience, please download the PDF above.

Field Guide

Deploying AI on Databases

Connect to databases, analyze and visualize with natural language to SQL

15+ Months of Client Operations

• Lessons
• Live Apps (4 Variants – 8 Apps)
• Source Code

GPT-5: Initial Assessment & Live Integration

Live App: [app.tigzig.com])(undefined)
DATS-4: Database AI Suite – Version 4

10ᵗʰ August 2025

Deployments

1. Build by Practitioner. Built for Business: The DATS-4 suite is built by a data scientist for internal teams. The design prioritizes analytical agility and rapid deployment in secure SMB environments. This involves different trade-offs than the standards for large-scale enterprise software.

2. Live History: The first client deployment was in April 2024. There are currently 9 live, customized versions running across 3 SMB clients.

3. Implementation Variants: Client projects vary based on need-specific components only, custom GPTs connected to databases, rapid-deploy version & customizations.

4. The Public App (app.tigzig.com): Fully functional version of the suite. It has been configured as a minimal security sandbox to allow for unrestricted testing of the core features.

5. Live Project Checklist: ALL client projects include a mandatory checklist: security layers, semantic model, fixed database connections, and disabling of admin features for end users

DATS-4 Evolution

Database AI Suite – Version 4

Field Report: GPT-5 First Look

From my experience, new model releases often have higher, volatile latencies and costs in the first days or weeks, then stabilize over time.

My preliminary assessment of GPT-5 first few days of release as of 10ᵗʰ Aug 2025:

1. Reasoning & Analysis: close to Claude Sonnet 4
2. Latencies: higher - temporary phenomenon
3. Costs: higher than expected. Likely temporary. On watch.
4. Variance: Given the amount of variance I am seeing with GPT-5, cost estimates would not be reliable. I am holding off on sharing exact cost estimates for GPT-5 for the time being. But based on published rates, I expect them to stabilize around GPT-4.1 levels
5. Integration: At the same time, I have incorporated GPT-5 as an LLM Choice options in the public DATS-4 for users to try out and compare results
6. Detailed cost comparisons and model choices are covered in the LLM section later in this guide

I typically migrate clients only once I am confident that the model performance, cost and latencies have stabilized.

1. Lessons

1. Security
2. Datamart & Context
3. Agent Setups
4. LLM Choices
5. LLM Cost
6. Usage Patterns
7. Platforms

Security

1. Align with rules – set by DB and server admins. They are troublesome but will save your bacon one day.
2. No end user touches the raw tables – even with SELECT access
3. Separate user ID’s at DB level with fine grained permissions
4. Row Level Security – use with Postgres
5. Separate: schemas / database / views for say Finance vs. Marketing. The additional maintenance effort is worth it.
6. Authentication: OAuth / API Keys
7. Log all API calls: push to a DB / 3ʳᵈ party tools
8. IP / Domain Whitelist: FastAPI / DBs / Agents / all end-points
9. CORS: for all FastAPI, with domain whitelist
10. Resource Limits for CPU & Memory – implement on server
11. Rate Limits: at FastAPI (with SlowAPI) and Agent end
12. Server: Firewalls, only SSH, Fail2Ban, IP Whitelists etc.
13. VPNs: default deployment always on a client VPN.

Security is expensive – direct cost, bandwidth and business opportunity loss. Every layer adds cost and user friction. Assess risk of breach for each data item, worse cases and potential impact. Apply layers accordingly. Everything is not catastrophic.

Building the Foundation

Datamarts and AI Context

Datamarts

Creating usable datamarts is one of the most time consuming things, especially the data cleaning and validating against reported numbers.

1. Need to know: create custom datamarts and views for specific use cases. Operate on need to know basis.
2. Auto Refresh: setup auto refresh of datamarts
3. Validation reports: validation reports for all datamart refreshes is mandatory
4. Be alert: After running for months, a validation can suddenly fail out of the blue. You must be ready to catch it.

Context

Use system prompt to provide context to AI

• Sample rows
• Univariates for numerics & distributions for categoricals
• Business rules and business context
• Golden queries – sample queries for common requests, particularly for the more complex queries
• Output formats / row limits / data gotchas

Agent Setup

Agent backend on Flowise AI, with conditional routing based on type of request. A multi-sequential agent setup

flowchart LR
    CR[Conditional Router Agent<br>gpt-4o-mini] -->|IF Advanced Analysis| AA[Advanced Analyst LLM Planner<br>Reasoning based analysis<br>plan + SQL queries<br>Choice of LLM]
    AA --> AAgent[Advanced Analyst Agent<br>Review, correct, execute, debug,<br>share analysis, charts & PDF output<br>Gpt 4.1]
    CR -->|IF Simple Request| GAgent[General Analyst Agent<br>Execute, debug,<br>share analysis,<br>results, charts,<br>PDF output<br>Gpt-4.1-mini]

Conditional routing agent will route to advanced analyst or general analyst based on a set of guidelines or if specifically instructed by user

Agent Setup

1. The Dispatcher (Conditional Router Agent)
This is the gatekeeper. Its only job is to analyze the user's request and route it to correct specialist agent based on a set of rules.

2. The Workhorse (General Analyst Agent)
This is GPT-4.1 mini - optimized for execution speed. It handles the majority of requests: direct SQL queries, data pulls, and standard charts. It does not perform multi-step reasoning. It directly executes, validates, and returns the result.

3. The Specialist (Advanced Analyst)
This is a two-step routing, used only for complex requests that require reasoning.
A. The Planner: First, a reasoning-focused LLM (choice of LLMs) creates a step-by-step analysis plan, including the exact SQL and python code required.
B. The Executor Agent: This is GPT-4.1 in all cases - reviews and executes that plan, performing final error checks and formatting the output.

The Executor agent will be upgraded to GPT-5 series once its cost and latency have stabilized

Equipping the Agents: Core Tools

The core DATS-4 agent uses three primary tools:

1. Database Connect

   • Custom FastAPI Server
   • Allows agent to connect to database to execute SQL queries

2. e2b Code Interpreter

   • Flowise built-in tool
   • Python sandbox
   • To create charts and run statistical analysis

3. Markdown to PDF

   • MCP Server
   • To create PDF (text only) output. The agent sends markdown to the MCP Server, which returns a PDF file path.

The system is modular, allowing other tools to be plugged in as needed: web scrapers, Excel updaters, report emailers, file converters, custom automations and more.

Agent Orchestration

To get an agent to deliver the right outcome, you have to test and calibrate—sometimes 100s of times.
It's the only way. These are the rules I follow:

1. No 100% : You will never get 100% what you instruct 100% of the time. Test and determine what variance you can live with.
2. Edge cases: Test edge cases and outliers. Calibrate instruction till you get your desired outcome.
3. Break it : Push it to limits. See where it trips and falls.
4. Reasoning required ? – if so, specify. Not always required.
5. Number of Queries – CRITICAL to specify a cap on number of SQL queries an agent can run for a single question.
6. CREATE / ALTER / DROP : Specify if they are allowed or not.
7. Temporary tables : Specify if permitted and how (CREATE TEMP or CREATE TABLE), and cleanup protocols.
8. Limit clause: How many rows? When to use? When not?
9. Division by zero: Common error – COALESCE(), NULLIF() etc.
10. Debug : Debugging protocol for query failures.
11. Reminders help – Remind to check for common issues – missing table, table exists, joins, data type mismatches etc.

Agent Backend

Don’t reinvent the wheel.

1. Don’t reinvent the wheel: use tools like Flowise/n8n as first choice - they take care of many nuances out-of-box. Connect user interface via API calls.

2. Flowise AI: is my first choice. Robust out-of-box memory and state management and numerous other features. Great for complex agent workflow, especially for sequential flows.

3. n8n – for app integrations and where Flowise not the best fit.

4. Hard-coded agents: used only for functionality that framework tools can't support.

LLM Choices

For end-user applications, use frontline providers (OpenAI, Google, Anthropic). They offer the best combination of reliability, consistency, quality and pricing. For internal analytics work – practitioners should test and use other models per their own judgment.

My Top Recommendations

1. SQL Executions: GPT-4.1 (GPT-5 once stable) for complex and 4.1-mini (GPT-5-mini once stable) for rest. GPT-4o-mini is excellent for simpler and repetitive requests.

2. Tool use: for all tool use functions, OpenAI’s GPT models - effective, reliable and cost efficient

3. Non-Tool LLM use: Gemini Flash 2.0/2.5 as first choice for non tool tasks - e.g. automations, schema detection, reasoning, planning

4. Complex: Claude Sonnet 4 for the hardest and most complex tasks

5. Other LLMs: DATS-4 provides LLM choices including DeepSeek, Qwen & GLM. Great quality and pricing. But I see a lot of variance in billed cost and latencies. DATS-4 allows for easy integration of other LLMs

LLM Costs: Guidelines

• Use Case: Always estimate for your specific use case. Check actual charged API costs. Don’t rely on published rate.
• Lowest Cost: GPT-4o mini and Gemini Flash 2.0 are older model, but robust, lowest cost and great for many tasks. Test them first
• Value: GPT-4.1-mini (GPT-5-mini once stable) and Gemini Flash 2.5 - great workhorses at reasonable cost.
• GPT-4.1 for harder tasks especially complex SQL executions (GPT-5 once stable)
• Claude Sonnet 4.0 is an all rounder and the best, but expensive. Keep for most complex reasoning.
• DeepSeek. Qwen, GLM and others – High latency and cost variance based on provider. DeepSeek more stable now.
• Single step agents for direct questions = low cost.
• Multi-step agents = exponential cost increase. Use with care. See next sections
• Number of SQL queries an agent is allowed has direct cost impact. 2 queries per question vs. 10 queries = 5x cost
• Context – piles up with same session adding to cost. New question = open new session

Cost estimates as of 10^th August 2025. Verify current rates before budgeting.

LLM Costs: 1 Question 1 Query

Cost Per 100 Simple Questions
1 Question = 1 SQL Queries / Tool Call
Single step: no reasoning step, direct execution

~USD per 100 Q

Example of single question

1. Share sample rows
2. Add new columns as per instructions
3. Join Table A & B by cust_id
4. Summarize by housing and show counts
5. Share chart for housing summary

• Actual vary by use case and the agent setup. Always estimate for your use cases and compare vs. actuals.
• For all estimates – keep in mind that as context increases the cost goes higher

Cost estimates as of 10ᵗʰ August 2025. Verify current rates before budgeting.

LLM Costs : Advanced Analysis

• In multi-step reasoning-based analysis, execution cost is biggest chunk due to multiple tool calls.
• 1 Question = 7-10 SQL Queries / Tool Calls.
• All executions by GPT-4.1. GPT-5 costs, once stable, likely to be around same levels.
• Time: ~2-3 mins per que. Can go upto 10m. Varies by question.

Advanced Analysis: Costs Vs. Quality Matrix

Claude Sonnet-4 and GPT-5 are top-tier models for advanced reasoning.

• Estimates based on live deployments & 250+ test runs
• Quality scores are a judgment-based assessment of the model's analytical reasoning depth
• Estimates vary – always estimate and check actuals for your use cases

*** GPT-5 is plotted at its projected stabilized cost (equal to GPT-4.1) for quality comparison only. Current costs are volatile and are not plotted.

Cost estimates as of 10ᵗʰ August 2025. Verify current rates before budgeting.

Warning: The Cost Multipliers of Multi-Step Agents

For multi-step analysis, costs don't just go up; they escalate exponentially - from 10X to 50X or more. This is a critical budget risk. Key factors:

1. No. of Steps:
   2 Step = double the context = 2X the cost

2. Number of tool calls
   Determined by # of SQL Queries allowed
   Simple Question = 1 SQL Query
   Analysis Question = 2 to 10+ SQL Queries
   =10X cost

3. Execution Model
   Needs stronger model GPT 4.1 vs 4.1 mini = 5X

4. Additional LLM Cost for reasoning
   ~ 7 cents per question for Sonnet 4

5. Context is a Multiplier, not an addition: larger semantic models, context, and system instructions don't just add to the cost; they multiply it with every step

6. Debugging
   In case of SQL query error - LLM will auto debug and re-run, taking up additional tool calls and costs.

Usage Patterns

The highest adoption I see is from operations, marketing, and finance teams. The following are the most common usage patterns from my client deployments

Operations, Marketing & Finance Teams

Natural language interface to backend datasets and uploaded CSV files

1. Pull specific customer and transactions records for review
2. Recon between finance and ops data
3. Insert / Update / Delete records
4. Download filtered data for offline analysis in Excel
5. Append fields and field cleanups
6. Generate summary reports with standard prompts for reuse
7. Generate PDF output

Many users prefer AI interface over their existing interfaces given the range of operations they can carry out and the efficiency of direct integration with automations

Analytics Folks

1. Pre-process raw tables and then download for offline analysis
2. Adhoc queries
3. Database level tasks requiring SQL

Platforms

1. Servers: Very often client determined. But where you have a choice, here are my defaults
   Server based: Hetzner + Coolify for deployments. Allows a firewalled environment to deploy any apps and databases. Reliable performance and pricing.
   Serverless : Vercel for React & Render / Railway for FastAPI

2. Databases

   • Neon: instant Postgres DB creation, deletion etc via API
     Top choice for AI apps requiring instant temp. databases
   • Aiven: great free tier.
   • Supabase: integrations esp. auth.
   • Standard / Self-Hosted: AWS RDS or Hetzner with Coolify

3. SQLite solid option for in-browser work. Requires setting up SQL Agent from the grounds up.

4. LLM Gateway: OpenRouter provides a single point gateway to all major LLM including the latest e.g GLM 4.5. Also, great set of reports to monitor costs.

5. Custom User Interfaces: React / NextJS / HTML-JS

6. Backend Services: FastAPI

2. Live Apps

DATS-4 : Deployed Live

[app.tigzig.com])(undefined)

Path: Database AI & SQL Apps

https://app.tigzig.com/landing-preview

TIGZIG: Co-Analyst
OSS | Analyze | Tools | GPTs | xlwings Lite | MCP | Voice

Search tools...

Recent Updates & Featured Apps

Database AI & SQL Apps

Complete Database AI Suite
Comprehensive database AI tools: DATS-4, ChatGPT connections, Flowise UI, Voice AI

• DATS-4 Database AI Suite
• ChatGPT Database Connections
• Flowise UI Rapid Deploy

xlwings Lite: Practice Lab

Learn xlwings Lite with Practical Examples
A Hands-on guide to get up and running fast with xlwings Lite

• AI-ready instructions & examples
• Step-by-step learning guides
• Interactive workbooks

Quants & Portfolio Apps

Run Portfolio Analytics
Quants and Portfolio analysis tools: Performance reports, technical analysis and more

• Security Performance Report
• AI Technical Analysis
• QuantStats Report

4 Variants – 8 Live Apps

All apps live, fully functional and open source.
DATS-4 is the flagship app.

Database AI and SQL Apps

A Suite of Open-Source Database Tools & SQL Agents for Practitioners

Full Suite

DATS-4 Database AI Suite
Connect to any DB, analyze TXT-CSV, advanced analytics with choice of multi-agent models

Custom GPT (ChatGPT) connected to Databases

Connect ChatGPT to any Database
Connect ChatGPT to any MySQL & PG database for querying and analysis and Python stats and chart

Connect ChatGPT to Supabase
Connect ChatGPT to Supabase for Natural language to SQL, charts, and Python stats

Connect ChatGPT to n8n
Connect ChatGPT to n8n for database queries, automation, Python, Google Apps Script

Rapid Deploy - Flowise

Adv. Analyst - Deepseek - Flowise UI
Advanced analytics powered by Deepseek R1 with Flowise UI. Connect to any Database

Quant + DB Analyst - Flowise UI
Quants, Technicals and Financials. PDF & Web Reports. Connect to any Databases. Flowise UI

Real time voice connected to Databases (Experimental)

Realtime Voice - ElevenLabs
ODI Cricket database analysis with python charts and stats. Eleven Labs voice widget connected to DB

RT Voice - OpenAI WebRTC
ODI Cricket database access. Uses OpenAI Realtime API WebRTC for queries and python charts and stats

[Link to app landing page: https://app.tigzig.com/database-landing]

DATS-4: How to Use

Option 1 : Customize & Deploy

Option 2: Try Live on app.tigzig.com

1. Customize & Deploy

Open Source
Customize & deploy on your server / VPNs

Core Deployment Patterns:

1. Full Suite: DATS-4 with custom UI.
2. Custom GPT: connected to the database backend.
3. Rapid UI: Flowise Agent UI for quick deployment.

Key Customization Areas

• Security
  • User API Keys / Oauth
  • Parameterized queries
  • DB user ID with restricted privileges
  • Row Level Security w/Postgres
• Context: Schemas, rows, queries, business rules
• Interface: customize based on user needs
• Components: deploy full suite or components
• Functionality: integrate additional functionalities

2. Try live on app.tigzig.com

The public site is a minimal-security sandbox configured to provide an unrestricted environment for testing the suite's full range of capabilities.

• Use on-the-fly temporary Postgres database generated by the app or create one instantly at Supabase / Neon / Aiven
• Use the sample files on the app / google drive

Practitioner’s Warning

WARNING: All database credentials and queries submitted via the public app are logged on the backend. Use this sandbox with non-sensitive data and credentials only.

ADMIN-LEVEL ACCESS: The full DATS-4 suite is an admin-level tool with extensive logging. For end-user deployment, you must restrict functionality and customize logging configurations.

Interface Components

• 2 Agents: Main Database Analyst and Quants Analyst
• Sample data for rapid testing
• Menu option to upload files and connect to databases
• On-the-fly temporary Postgres database
• Choice of LLM for advanced analysis
• Chart & Document pane
• Logs
• File uploads: interactive grid and automated data quality metrics

DATS-4 Database AI Suite | Connect to Any Database | Text2SQL | Advanced Reasoning | Python Charts & Stats | OpenAI Claud

Menu | Gemini 2.0 Fla.. (Model) | Try Sample

FILES

• Try Sample
• Choose File
• Fast Insights

DATABASE

• Connect to DB
• Upload File to DB
• Export Data
• Create New DB

SETTINGS

• Hide Charts
• Logout

Tabs:
Quick Insights | AI Quants Analyst | AI Database Analyst | Logs | Charts

Start a conversation by typing a message below.

Docs:
GENAI LLM
GENAI APP H

Choice of LLM for Advanced Analysis

Choose your LLM for the reasoning step. The app setup also allows an efficient method to add and remove LLMs

Cost varies based on use case and AI setup. Refer attached price study for more details.

[Open Price Guide]

[Cancel] [x Confirm Selection]

Core Workflows

1. Sample File Test
   Use the built-in sample data and an on-the-fly temporary database for rapid evaluation

2. File Upload
   Upload a local file (CSV/Tab Delimited) to a temporary or user-provided database.

3. Direct Connection
   Connect directly to a remote Postgres or MySQL database

1. Sample File – Rapid Test

Menu    Gemini 2.0 Fla...  Model    [Try Sample]    Ex  Ex

2. Choose Sample Dataset

This is Reserve Bank of India's monthly statistics report on ATMs, credit cards, debit cards, and POS transactions. The dataset provided is for December 2024.

3. Database Connection Required

Please select how you would like to proceed with the data analysis:

Use on-the-fly temporary Postgres database OR connect to your own DB.

4. Copy and paste ready to use starter prompt for quick analysis once database is setup. Or go with your own request.

2. Upload Your File

• Select your file for upload. Supports CSV and tab delimited
  • Menu
  • Gemini 2.0 Fla...
  • FILES
  • Try Sample
  • Choose File

• What would you like to do?
  • DATABASE
  • Upload to Database

• Use Temporary Database
  • Recommended for quick analysis. Creates an instant database session.
• Connect Own Database
  • Connect to your existing database or create a new DB

4. File schema sent to AI automatically. Go to Advanced Analyst tab and ask questions, analyze, create and customize charts – in natural language

3. Connect your Database

Postgres and MySQL supported

1. DATABASE
   Connect to DB

Menu -> Connect to DB

2. Connect to Database

| Connection Details * | Paste your connection details here... |

Paste your DB credentials. Format does not matter – URI / table / text – AI will parse it

• Go to Advanced Analyst tab
• The AI agent does not automatically know your database schema upon connection. You must instruct it to list tables or query sample rows to provide it with the necessary context for analysis.
• Query, Analyze, Merge, Summarize, Visualize

Agent Reasoning View

Full trace of agent’s reasoning process from business context and feature engineering to the final SQL queries and debug logs

Advanced Analyst

Flow Step: IlmAgentflow_0

Reasoning Process

Business Context & Analytical Goal:

The objective is to build a customer profiling data mart for advanced analytics and machine learning.

1. One record per customer by merging demographic data (BANK_UNQ_1M) with transaction data (TRANS_10M).

2. Focus on transaction behavior: Differentiate between cash (TC_CODE=1001) and other transactions.

3. Derived features: Create up to 20 insightful variables summarizing transaction behavior (e.g., averages, ratios, recency, frequency).

Key Considerations:

• Data Volume: TRANS_10M has 10M+ rows. Aggregations must be efficient.

• Feature Engineering: Prioritize variables that capture:

  • Monetary behavior: Total/average spending per transaction type.
  • Behavioral ratios: Cash vs. retail preferences.
  • Temporal patterns: Recency (last transaction), frequency (activity span).
  • Variability: Standard deviation of transaction amounts.

• Technical Constraints:

  • Use permanent tables (no temporary tables).
  • Limit to 6 SQL queries.

SQL Queries (MySQL)

Analysis Queries (Data Mart Creation)

Query 1: Create Intermediate Transaction Aggregation Table

CREATE TABLE IF NOT EXISTS trans_agg_temp AS
SELECT
    CUSTID,
    SUM(CASE WHEN TC_CODE = 1001 THEN AMOUNT ELSE 0 END) AS cash_amount,
    SUM(CASE WHEN TC_CODE = 1001 THEN 1 ELSE 0 END) AS cash_count,
    AVG(CASE WHEN TC_CODE = 1001 THEN AMOUNT ELSE NULL END) AS cash_avg_amount
FROM TRANS_10M
GROUP BY CUSTID;

Dedicated Charts & Docs Panel

Model Try Sample [ ] [ ]

• Single-click toggle to open/close charts & document panel
• Dedicated, full-screen chart panel for visualizations.
• Dynamic document panel for live report and data updates.

Charts

  </td>
</tr>

Docs

About · Privacy · Terms

Python Charts

Integrated Python Interpreter for charting. All charts below were generated directly in the app.

Statistical Analysis

The integrated Python Interpreter enables full statistical analysis, not just charting.

Chi-Square Test of Independence: Education vs Job

Statistical Output Table

Interpretation (Compact)

There is a statistically significant association between education and job (χ² = 531,242.61, df = 20, p < 0.0001). This means the distribution of job types is not independent of education level among customers.

ANOVA Test: Average Transaction Value by Education Level

ANOVA Summary Table

Group Comparison Table

Interpretation

There is a statistically significant difference in average transaction value across education groups (p < 0.05).

PDF Output

• On-demand, formatted PDF report generation for all analysis and query outputs (text only)
• Report structure and content are fully customizable via natural language instructions.

RBI Credit Card Segment: Top Bank Rankings & Insights (March 2025)

Analysis Objective

To identify and rank the top 10 banks in India’s credit card segment for March 2025, using a composite scoring system based on card issuance, transaction volumes, values, and usage intensity. The goal is to provide actionable insights into which banks are leading in both scale and customer engagement in the credit card market.

Methodology

1. Data Preparation & Derived Metrics

• Source: RBI monthly statistics for March 2025, table: rbicc_mar2025_78680
• Key Metrics Derived per Bank:
  • credit_cards_nos: Number of credit cards issued.
  • total_txn_volume: Total credit card transaction count (POS + E-com + Others).
  • total_txn_value: Total credit card transaction value (POS + E-com + Others).
  • avg_txn_value: Average value per transaction = total_txn_value / total_txn_volume.
  • usage_ratio: Transaction volume per card = total_txn_volume / credit_cards_nos.

rbicc_mar2025_78680

2. Normalization

• Each metric is normalized using min-max scaling to [0,1] across all banks:
  $$ \text{norm_metric} = \frac{\text{value} - \text{min}}{\text{max} - \text{min}} $$

3. Weighted Scoring System

• Weights assigned:
  • Credit cards issued: 25%
  • Transaction volume: 35%

Detailed Logs

• Detailed logging of key API calls and actions
• Valuable for first line of debugging

Tabs:
AI Data Structure | AI Quick Insights | AI Quants Analyst | AI Database Analyst | Logs

Auto-scroll: Off

2025-08-07T03:41:42.864Z
Making OpenAI API request via proxy server...

2025-08-07T03:41:42.864Z
Using RT endpoint: https://rtephemeral.hosting.tigzig.com

2025-08-07T03:41:42.864Z
Request Body: {
  "model": "gpt-4o",
  "messages": [
    {
      "role": "system",
      "content": "You are a specialized database credentials parser. Return only valid JSON without any markdown formatting or additional text."
    },
    {
      "role": "user",
      "content": "You are a specialized database credentials parser. Parse the following connection details into a standardized JSON format.\n\nBackground:\n- These credentials will be used for automated database connections\n- The format must be exact as it will be used directly in code\n- All values must be strings\n- The response must be valid JSON without any markdown or additional text\n\nRequired Fields (all must be present):\n1. host: The database server hostname/IP\n2. database: The database name\n3. user: The username for authentication\n4. password: The password for authentication\n5. schema: The database schema (default to \"public\" if not specified)\n6. port: The connection port (use defaults if not specified)\n7. db_type: Must be either \"postgresql\" or \"mysql\"\n\nRules:\n- Default Ports:\n  - PostgreSQL: use \"5432\"\n  - MySQL: use \"3306\"\n"
    }
  ]
}

Full logs are ‘admin’ level with sensitive info being logged. Restrict as per security access. For end users, remove / customize logging as per requirement.

Export Tables

• Perform transformations and create new tables, then export any table to a local file (CSV or Pipe Delimited).
• Full support for both MySQL and Postgres environments.

Interactive Data Table & DQ Report

Data Table

Statistics for ATM_CRM_ONSITE_NOS

Calculations under validation & enhancement

Central Tendency

Distribution Plot

(Box plot representation, not shown here)

Percentile Distribution

Dispersion

Distribution Shape

Position

Percentiles

Data Quality

Additional Information for BANK OF BARODA

• Interactive data grid for all uploaded files.
• On-the-fly descriptive statistics and data quality assessment.
• Record-level view with filtering and sorting capabilities.

SSO with OAuth

Welcome
Log in to DATS-4: Database AI Suite to continue to DATS-4.

• OAuth-based single sign-on (SSO) via Auth0 for - Google, LinkedIn, Microsoft, GitHub and Amazon.
• Current Scope: The baseline implementation is linked to ‘Create DB’ function only. This provides unrestricted testing of analysis tools without forcing an app level login.

This baseline setup is built for extension. It provides the OAuth foundation needed for full app authentication, fine-grained access controls, and row-level security in a live client project.

App Variants

There are three stable variants of the app, each suitable for a different use case. The fourth, Voice AI, is experimental.

1. Custom GPT

• Custom GPT connected to databases is a robust, effective solution - straightforward setup and low maintenance
• Combines a front-end UI, built-in AI Agent, and the full native ChatGPT feature set
• No separate API Cost for Agent + GPT-5 access
• Efficient to connect automation backends and other apps via FastAPI / n8n / Flowise / Make.com
• This is my first choice

2. Flowise / n8n

• Built-in user interfaces from Flowise and n8n.
• Setup is efficient with direct connection to automation backends and other apps
• API Charges - as per usage

3. Database AI Suite -4

• Top choice where full feature and customization needed
• Fully customisable: user interface as well as backend
• Deploy anywhere, connect to Oauth
• API Charges - as per usage

3. Source Code

Source Code

All source codes links on app page in docs section

• Hit Docs on the app page
• Links to source codes and build guide, including video guides

Docs Section Screenshot

• The screenshot shows a highlighted Docs button on the app page.

• Under Docs, there are two tabs: Quick Start and Build It.

• The Build Guide section states:

  The full app has 7 major components:

  1. Main App: The main application with the UI
  2. FastAPI Server: Database Connector: Handles Text-to-SQL processing, including file uploads.
  3. FastAPI Server: Neon DB Creation: Handles Text-to-SQL processing, including file uploads.

GitHub Repositories

1. Main Repo
   https://github.com/amararun/shared-rexdb-auth-embed-v3-agentflowv2

2. FastAPI Server : SQL DB Connect
   https://github.com/amararun/shared-fastapi-rex-db-coolify

3. FastAPI Server: Neon Database
   (URL not fully visible in screenshot)

Architecture

Modular architecture for efficient integration of automation services or backend

      </td>
      <td></td>
    </tr>

      </td>
      <td>
        

      </td>
      <td>
        

      </td>
    </tr>

</td>

Components Based Architecture

• Mix and match deployment of individual tools
• Connect components to your own user interface

Numerous more components available open source at [app.tigzig.com])(undefined) - web scraper, pre-formatted slide deck creator, Excel table to PDF, Excel and Google Sheets updater, file converters, finance data extractors…

Architecture Overview Doc

This is my personal app architecture file for DATS-4 that I feed to AI Coder at start of every session. Enables immediate productivity without full codebase exploration. Includes critical gotchas from earlier experiences.

DATS-4 - AI Coder Documentation

🏗 Architecture Overview

Tech Stack: React 18 + TypeScript + Vite + TailwindCSS
Deployment: Vercel (can run in iframe or standalone)
Authentication: Auth0 (iframe limitations handled)
Data Processing: File upload → AI analysis → Database storage → Visualization

📦 Key Dependencies

Core: react, typescript, vite, tailwindcss
UI: @radix-ui/* (dialogs, dropdowns, tooltips), lucide-react (icons)
Data: ag-grid-react (tables), axios (HTTP), uuid
Auth: @auth0/auth0-react
Styling: tailwindcss-animate, class-variance-authority, clsx
Markdown: react-markdown, react-syntax-highlighter

📁 Project Structure

frontend/src/
├── App.tsx                          # 🔥 MAIN APP - All state & orchestration (2094 lines)
├── components/
│   ├── ui/                         # Shadcn-style base components
│   ├── file-upload-section.tsx     # File selection handling
│   ├── analysis-tabs.tsx           # Main tabbed interface

README

All GitHub repos with README with step-by-step guide

DATS-4 Database AI Suite

Live App

A full version of this app is deployed and available at app.tigzig.com

Documentation

• User Guide - Comprehensive guide on how to use the application
• App Architecture - Detailed explanation of the application architecture
• Setup Guide - Continue reading below for installation and setup instructions

About The Project

The DATS-4 Database AI Suite is a tool for AI-driven data analysis against relational databases. Its primary function is to connect to remote PostgreSQL or MySQL databases to execute queries. Queries

Getting Started

Follow these instructions to set up and run the project locally:

1. Clone the Repository
First, clone the repository to your local machine using the following command:

git clone <https://github.com/amararun/shared-rexdb-file-upload> .

Note: Check for the latest shared repository name in case it has changed.

2. A Note on Analytics Tracking

TIGZIG: Micro-Apps for Analytics

25+ apps: Database AI / xlwings Lite / Automation / Quants

Access the full suite of open-source tools at

[app.tigzig.com])(undefined)

xlwings Lite: Practice Lab NEW
Learn xlwings Lite with Practical Examples
A Hands-on guide to get up and running fast with xlwings Lite

Quants Suite NEW
Analyze your portfolio performance
Combined QuantStats, Security Performance, AI Powered Technical Analysis, Financials and Prices

DATS-4 Database AI Suite NEW
Database AI Suite - V4
Connect to any DB, analyze TXT-CSV, advanced analytics with choice of multi-agent models

TIGZIG Quants Agent [NEW]
AI Assisted Quants Reports
Portfolio stats, AI Powered Technical Chart Analysis & Yahoo Finance extraction. PDF & Web Reports

Adv. Analyst - Deepseek - Flowise UI
Advanced analytics with Deepseek R1
Advanced analytics powered by Deepseek R1 with Flowise UI. Connect to any Database

Quant + DB Analyst - Flowise UI
AI-powered portfolio analysis
Quants, Technicals and Financials. PDF & Web Reports. Connect to any Databases. Flowise UI

AI Powered MF Processor [NEW] NEW
Convert MF portfolio files
Convert MF portfolio disclosures files from excel to CSV/TXT

Convert PDF to Text with Llama Parse
Advanced PDF conversion
Advanced PDF to text conversion including complex tables and layouts. Powered by

Convert any File to Text Markitdown
Universal file converter
PDFs, Excel, Word, PPT, etc to text format suitable for LLM input. Powered by Microsoft

Amar Harolikar
Specialist - Decision Sciences & Applied AI
Builder of app.tigzig.com