SShortSingh.
Back to feed

BIAN series: which team owns each step of a bank account opening

0
·3 views

A new installment of the BIAN and banking architecture series follows a retail account opening journey to show where each responsibility starts and ends. The example describes a customer whose app reports approval while the account remains unusable, leaving staff unsure whether the cause is a pending check, an expired offer, or central banking processing. The author proposes splitting the journey into channel, eligibility, offer, and account-opening responsibilities, each with clear contracts and its own identifiers. The approach draws on BIAN's Service Landscape and its Customer Offer and Current Account references in version 14.0. The aim is to let different teams explain and recover a case stalled between steps, in a hypothetical bank aiming to cut the opening time from five business days to one.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Oronode Lumina Aims to Be a Full Local AI Desktop Environment

Oronode Lumina is presented as a desktop environment that combines local AI with coding, automation, creative tools, and everyday workflows rather than a single-purpose AI tool. The platform lets users run AI models directly on their own machines and work with models from different providers. It also covers AI-assisted development, image, video, and audio creation, and computer automation alongside document, spreadsheet, and presentation handling. Integrations and plugins are included to extend its capabilities, with the aim of making AI a broader working environment rather than just a chatbot. The announcement was shared on DEV Community, where the author invited developers to suggest what a complete local AI environment should include.

0
ProgrammingDEV Community ·

Tutorial Shows How to Trace Vulnerable Dependency Versions Across Releases with GitHub Actions

A walkthrough published on DEV Community explains how to attach a software bill of materials (SBOM) to each GitHub release and check those SBOMs against a vulnerable version range. The demonstration uses a fictional advisory for the is-number package, with an affected range of 6.0.0 and above. Under the example workflow, release v1.2.0 is flagged as containing the affected version, while v1.3.0, which ships 7.0.0, falls outside the range. The tutorial also covers running the comparison in a GitHub Actions workflow that writes results to the job summary and fails if an SBOM is missing or unreadable. It links the approach to the EU Cyber Resilience Act, whose SBOM requirements apply from December 2027 and whose reporting duties began in September 2026.

0
ProgrammingDEV Community ·

Developer's Oct 9 build day: obsidian stone site, sound redesign and live AI preview

On October 9, a developer launched a new domain at 3:16 am and spent the day building a site whose main visual is a chipped obsidian stone that responds to cursor or finger movement. Early sound attempts, including bells, a glass rub and chipmunk-like noises, were rejected, and the developer settled on removing everything except the tap sound. The AI tool used for the work admitted it could not judge how its audio sounded, so the developer made the final decisions. The same day, the Enter button was reworked so it responds instantly while content loads, and a live AI preview was added that generates a sample site direction from a typed business description. A footpath tool on a separate project was also changed to find roads automatically rather than requiring manual clicks.

0
ProgrammingDEV Community ·

Teaching a Small LLM a New Character Erased Its Old Stories: Perplexity Rose From 9 to 156

A tutorial author fine-tuned a 5.9-million-parameter story model, trained on bedtime stories about Momo the monkey, to learn a new character, Zuri the zebra. Training on 300 new Zuri stories took about fifteen seconds of effort but made the model roughly seventeen times worse on its original stories, a failure known as catastrophic forgetting. The author explains that no memories are deleted; the same model weights are repeatedly adjusted toward the new task, so the old capability gradually degrades. To measure this, the author tracks perplexity, where lower is better, on the old stories. The post also describes a remedy being tested, replaying a share of original training windows during fine-tuning, and compares a run without replay against one using 10% replay.