Back to RegIntel Home
SA
Sumit Arora

Full-Stack Architect

Brisbane, Australia
March 2026
12 min readBusiness SolutionRegIntel — Part 5 of 6

Part 5 — The Team & How It Delivers

Solution Architect, 4 engineering disciplines, BA, PM, and domain experts. What each persona actually receives from the system. How the team goes from discovery to production.

1

The Full Team — Who Is Involved and What They Own

Click any role to see full responsibilities, tools, and how they connect to other disciplines

A note on team size and founder skill

This is how a cross-functional team would divide this work in an industry setting. In startups, a skilled founder or small team of generalists can cover multiple layers — and often does. The real skill of a founder is understanding the problem deeply enough to know which layers exist, how they connect, and when to get help. The layers do not disappear because the headcount is small — someone still owns each one.

Business Analyst

Translates domain rules into structured requirements. Works with lawyers, CAs, and compliance officers to validate extractions. Writes acceptance criteria for the knowledge base.

Product Manager

Owns the product roadmap. Prioritises which domains and personas are built first. Defines what "good enough" means for confidence scoring. Bridges engineering decisions to customer value.

Domain Experts

CAs, compliance officers, regulatory lawyers who validate extractions and sign off on the knowledge base. Without expert review, AI extraction accuracy on complex regulatory text is 85–92%. The remaining 8–15% contains the cases that matter most.

2

What Gets Delivered — Persona by Persona

Click any persona to see the full 6-step journey from query to answer — layer by layer

3

How the Team Actually Gets Work Done

Tools, process, and the development lifecycle from first commit to production

When multiple engineers work together, you need shared tools to coordinate. At GetPost Labs we use a standard modern development process. If you have never worked in a team before, this is the operating model.

JIRA

Work Item Tracking

Every piece of work — a feature, a bug fix, a research spike — becomes a ticket in JIRA. Tickets are assigned to team members, estimated, and moved through stages: To Do → In Progress → In Review → Done. The Solution Architect and PM use JIRA to see what the whole team is working on at any moment.

Watch: JIRA for beginners

Confluence

Knowledge Documentation

Confluence is where the team writes down decisions, designs, and how-to guides. Every architecture decision, API spec, and onboarding guide lives here. When you join a project, Confluence is where you go to understand what has already been decided and why.

Watch: Confluence intro

GitHub + CI/CD

Code & Deployment Pipeline

All code lives in GitHub. When a developer pushes code, a CI/CD pipeline automatically runs tests, builds a Docker image, and deploys it to the right environment. CI = Continuous Integration (merging code frequently). CD = Continuous Deployment (shipping to production automatically after tests pass).

Watch: CI/CD explained

The Development Lifecycle — From Idea to Production

01

Discovery & Requirements

BA + PM + Domain Expert

The Product Manager defines what needs to be built. The Business Analyst translates domain rules into structured JIRA tickets. Domain experts validate that requirements are legally correct.

02

Solution Architecture

Solution Architect

The Solution Architect designs how all components connect. Architecture Decision Records (ADRs) are written in Confluence. API contracts that other teams will build to are defined here — the shared blueprint.

03

Schema & Data Pipeline

Data Science

DS builds the document registry schema in PostgreSQL, sets up the ingestion pipeline in Airflow (MWAA), and begins loading documents into S3. DS owns the ground truth of what documents exist and their metadata.

04

Extraction & Retrieval

AI / LLM Engineering

AI/LLM builds the extraction pipeline — chunking, obligation extraction, embedding. They build the GraphRAG retrieval layer on top of the structured corpus. Evaluation experiments measure answer quality.

05

Infrastructure & CI/CD

Platform Engineering

Platform writes Terraform for all AWS infrastructure. GitHub Actions CI/CD is configured so any team pushing code is automatically tested and deployed to the dev environment. CloudWatch dashboards go live.

06

API & Frontend

Full-Stack Engineering

Full-Stack builds the REST API, query interface, provenance display, and admin dashboard — built against the API contract defined by the Solution Architect in Step 02.

07

Testing & Expert Review

BA + Domain Expert + QA

The BA runs the knowledge base through domain expert review. CAs and compliance officers check AI-extracted obligations for accuracy. Bugs go back to JIRA. Nothing goes to production without sign-off.

08

Deployment to Production

Platform Engineering + Solution Architect

Platform runs the production deployment through the CI/CD pipeline. The Solution Architect signs off. CloudWatch monitoring goes live. Incidents go to an on-call runbook.

Want to explore what you can build or achieve?

Whether it is a product idea, a compliance challenge, or an engineering question — let's talk through it.