Build Financial Software with Generative AI (From Scratch)

brief contents....6

contents....7

preface....14

acknowledgments....15

about this book....16

Who should read this book....16

How this book is organized: A road map....16

About the code....17

liveBook discussion forum....18

about the authors....19

about the cover illustration....20

Part 1....21

1 Exploring FinTech and generative AI....23

1.1 The Futuristic FinTech project....24

1.2 ACH dashboard....25

1.2.1 How important is the ACH?....26

1.3 SAFe Agile....27

1.4 The ACH dashboard....29

1.4.1 Tools of the trade....33

1.5 A first look at generative AI....33

1.5.1 A better rubber duck....34

1.5.2 Hello, Check 21!....34

1.5.3 Help! This code is not working!....38

1.5.4 Help! I forgot the syntax!....41

1.5.5 Did I just give away my data?....44

1.5.6 Privacy and security....46

1.5.7 Staying up to date....46

2 Parsing payments....49

2.1 Modernizing our legacy software....49

2.2 Understanding the ACH....51

2.3 Parsing an ACH file....52

2.3.1 Asking ChatGPT to parse an ACH file....54

2.3.2 Parsing an ACH file with Copilot....58

2.3.3 Generative AI: Trust but verify....60

2.4 Automated testing....61

2.4.1 Testing the number of records read....62

2.4.2 Parsing the records....62

2.4.3 File header record (type 1)....65

2.4.4 Batch header record (type 5)....68

2.5 Entry detail record (type 6)....70

2.5.1 Addenda record (type 7)....71

2.5.2 Batch control record (type 8)....72

2.5.3 File trailer record (type 9)....73

2.5.4 Passed!....74

2.6 The not-so-happy path....74

2.6.1 File rejection....75

2.6.2 Batch rejection....76

2.6.3 Entry rejection....77

2.7 Interpreting the code....77

2.7.1 Wheres my switch statement?....77

2.7.2 Type hints....78

2.7.3 Secure coding....79

Part 2....83

3 Getting started with Docker....85

3.1 Where to begin....86

3.2 Creating a docker-compose.yml file....88

3.2.1 Creating a CloudBeaver container....90

3.2.2 Create a PostgreSQL container....92

3.2.3 API container....94

3.2.4 Web server container....96

3.3 Connecting our containers....96

3.3.1 Connecting to the database....97

3.3.2 Troubleshooting PostgreSQL and FastAPI....98

3.4 Calling our API....99

3.4.1 Troubleshooting our web site....102

3.5 Container security....103

3.6 Optimizing Docker....105

3.6.1 General maintenance....106

3.6.2 Optimizing image size....107

3.6.3 Optimizing build time....108

3.7 Removing the hardcoded usernamepassword....109

3.8 Health checks....112

4 APIs: The piping between our components....114

4.1 API-driven design....114

4.2 Our APIs....116

4.3 Building our APIs with FastAPI....117

4.3.1 Retrieving files with GET files....119

4.3.2 Adding a file with POST files....122

4.3.3 Finishing the rest of the APIs....124

4.4 Adding our API to Docker....124

4.4.1 Preparing the project....125

4.4.2 Docker setup....125

4.5 Documenting our API....126

4.5.1 Swagger....127

4.5.2 Redoc....128

4.5.3 What is so great about API documentation?....128

4.6 Managing our APIs....129

4.6.1 Benefits of API management....129

4.6.2 Adding WSO2 API manager to our project....130

4.6.3 OpenAPI Version....131

4.6.4 Create an API in WSO2....132

4.6.5 Troubleshooting....133

4.7 Testing with Postman....134

4.8 Load testing with Locust....136

4.9 Improving our code....141

4.9.1 Using a prefix and versioning....141

4.9.2 Adding Pydantic....142

4.9.3 Validation with Pydantic....143

4.9.4 Troubleshooting Pydantic....145

5 Storing our ACH files....147

5.1 Designing our database....148

5.2 Using SQL directly....152

5.2.1 Adding records to the ach_files table....153

5.3 Storing the file header record....158

5.3.1 Using generative AI....158

5.3.2 Full example....159

5.4 Storing the rest of our ACH records....163

5.4.1 Storing ACH files challenge: Lessons learned....163

5.5 Storing exceptions....165

5.6 Uploading an ACH file....167

5.7 Storing records with integrity....172

5.8 Using Pydantic....173

5.9 Lessons learned....176

5.10 Coding changes....179

5.10.1 Creating Pydantic schema for the unparsed ACH records....179

5.10.2 Creating Pydantic schema for the parsed ACH records....179

5.10.3 Unit test changes....180

5.11 Design and different approaches....180

6 Taking the next step with Next.js....183

6.1 Getting started with Next.js....184

6.2 Creating an upload page....186

6.2.1 Testing our upload page....186

6.3 Refactoring our navigation....188

6.4 Converting recent order to recent ACH uploads....190

6.5 Mocking our ACH uploads data....195

6.5.1 Adding WireMock to Docker....196

6.5.2 Setting up mappings in WireMock....197

6.5.3 Accessing endpoints with axios....199

6.6 Expanding the UI with an ACH batch view....201

6.6.1 Defining our WireMock data for the ACH batch view....202

6.6.2 Updating the dashboard to take us to the batches screen....204

6.6.3 Displaying batch details....205

6.7 Creating a ScatterPlot component....206

6.7.1 Scatterplot component....207

6.7.2 Scatterplot custom tooltip....209

6.7.3 Scatterplot custom ticks....209

6.7.4 Using colors for the chart....210

6.8 Deploying the dashboard in Docker....212

7 Our minimum viable product....214

7.1 Which minimum are we talking about?....215

7.2 Preparing for the MVP....216

7.3 Building out the files APIs....217

7.3.1 Testing the API....224

7.4 ACH batches APIs....227

7.4.1 Adding the filesidbatches API....227

7.4.2 Supporting ACH batches in the dashboard....233

7.4.3 Uploading files....235

7.5 Putting the puzzle together....237

7.5.1 Cleaning up the configuration files....237

7.5.2 Ensuring APIs are accessible....240

7.5.3 Programming challenges....240

7.5.4 Exploring the dashboard....241

7.6 Gotta have more files....241

7.7 Adding a basic login page....244

7.8 Adding TLS to our dashboard....249

7.9 Testing with Playwright....251

7.10 MVP wrap-up....253

7.10.1 Troubleshooting the MVP....253

Part 3....257

8 Exceptions....259

8.1 Planning enhancements....260

8.2 Exceptions and warnings....261

8.3 Creating ACH files with exceptions....264

8.3.1 Creating ACH files with file-level exceptions....264

8.3.2 Creating ACH files with batch-level exceptions....265

8.3.3 Creating ACH files with entry-level exceptions....266

8.3.4 Clean-up of our file creation syntax....266

8.4 Expanding the ACH processing for exceptions....267

8.4.1 ACH exceptions in the database....267

8.4.2 Recovery options....269

8.5 Updating the ACH parser....270

8.5.1 Creating unit tests for the exceptions....271

8.5.2 Supporting the exceptions in our parser....272

8.5.3 Validation in Pydantic....275

8.6 Updating the APIs....276

8.6.1 Retrieving an unparsed record....282

8.7 Insights into exceptions with the UI....283

8.7.1 Showing all the exceptions....284

8.7.2 A visual cue for exceptions....287

8.8 Additional UI changes....289

8.8.1 Creating an exceptions component....289

8.8.2 Batch statistics component....291

8.8.3 Adding the application....295

8.8.4 Making our UI more responsive....297

9 Searching and auditing....301

9.1 Sprint planning....302

9.2 Searching ACH transactions....303

9.2.1 BDD testing for searching....304

9.2.2 Search API....306

9.2.3 UI search page....311

9.2.4 Integration testing with Playwright....315

9.3 Auditing user interactions....317

9.3.1 Database design....317

9.3.2 TDD testing for auditing....318

9.3.3 Auditing logic....320

9.4 Using middleware for our logging....321

9.5 Viewing the audit log....323

9.5.1 Creating the page....323

9.5.2 Creating the page components....324

9.5.3 Adding the APIs for the audit log....328

10 Company information....331

10.1 Sprint planning....332

10.2 Company Information....333

10.2.1 Database design: Company information....333

10.2.2 Using generative AI to generate mock data ....337

10.2.3 API design ....339

10.2.4 User interface....342

10.3 Company dashboard....346

10.3.1 Navigation for the company dashboard....346

10.3.2 Creating ACH files for our companies....347

10.4 Company Information component....348

10.5 Recent Batches component....353

10.5.1 Implementing the recent batches API....353

10.5.2 Displaying recent batches....356

10.6 Expected Files component....357

10.6.1 Storing expected files....358

10.6.2 Implementing the expected files API....359

10.6.3 Displaying the expected files....363

10.7 Company Limits component....364

10.7.1 Storing the company limits....365

10.7.2 Implementing the company limits API....366

10.7.3 Displaying the company limits....369

11 International ACH transactions and OFAC scanning....375

11.1 Sprint planning....376

11.2 IAT....377

11.2.1 IAT batches: An overview....377

11.3 Creating IAT batches....379

11.3.1 Database changes....381

11.4 IAT record parsing....383

11.5 IAT file processing....387

11.5.1 Unit testing....388

11.5.2 Updating file processing....392

11.6 Effects on the dashboard....395

11.6.1 The get_batches method....395

11.6.2 Batch entries....396

11.7 OFAC scanning....398

11.7.1 Sanctioned individuals and countries....398

11.7.2 Scanning for individuals....399

11.8 API....403

11.8.1 Unit test....404

11.8.2 Creating the endpoint....405

11.8.3 Finishing the API....405

11.9 User interface....407

11.9.1 The OFAC page....407

11.9.2 OFAC components....408

11.9.3 OFAC results....410

12 Where to go from here....413

12.1 Backend processing....414

12.1.1 Asynchronous processing....414

12.1.2 ORM frameworks....415

12.1.3 APIs....415

12.2 Infrastructure....416

12.2.1 Docker....416

12.2.2 Multitenant environment....416

12.2.3 Continuous integration and delivery....417

12.2.4 Testing....417

12.2.5 Database....417

12.3 User interface....418

12.3.1 Charts and forms....418

12.3.2 Configurable dashboards....420

12.3.3 Mobile version....420

12.4 Finance....420

12.4.1 Accounts and customers....420

12.4.2 ACH origination and returns....421

12.4.3 Positive Pay....422

12.4.4 Service charges....422

12.5 Final thoughts....422

index....424

Build working and regulation-compliant financial software—from scratch!

The software used by banks, trading firms, and other financial services has special requirements at every level, from securing the UI to making sure backend services comply with a host of regulations.Build Financial Software with Generative AI (From Scratch)shows you how to deliver full stack financial services software—and how generative AI can make you even more productive.

In Build Financial Software with Generative AI (From Scratch) you will:

• Explore the core concepts of FinTech
• Speed development with generative AI tools
• Develop and deploy containerized services
• Create and document APIs
• Effectively visualize your data

In Build Financial Software with Generative AI (From Scratch) you’ll build working software for processing Automated Clearing House (ACH) files, a cornerstone technology of banking that moves trillions of dollars every year. You’ll work with generative AI technology throughout the full stack application, including researching the tech for your application, spinning up a bare bone starting project, answering domain questions, clarifying functionality, and troubleshooting. Along the way, you’ll learn what sets FinTech projects apart from normal web apps.

About the technology

The financial industry is awash with regulatory and compliance challenges, complex technical requirements, and stringent security demands. There’s a huge demand for developers who can create financial services software and this book will get you started. You’ll build your own FinTech app from the ground up—with a big productivity boost from Generative AI!

About the book

Build Financial Software with Generative AI (From Scratch) guides you through modernizing a full-stack Automated Clearing House (ACH) application, layer-by-layer. You’ll start with a quick review of FinTech basics and an introduction to GenAI tools. Then, you’ll develop a data visualization dashboard with React, containerize components with Docker, create and refine APIs, implement backend processing, and even design a custom database. Throughout, you’ll see how AI tools aid with coding, testing, research, security, documentation, and even Agile practices.

What's inside

• Learn the core concepts of FinTech development
• Create and document APIs using Generative AI
• Build an awesome data visualization dashboard

About the reader

Examples are in Python. No experience with generative AI or financial services required.