Skip to content

paulasilvatech/hackathon-datacorp-team-kit

Repository files navigation

title Team Repository Kit
description Starter kit for hackathon team repositories with narrative context, personas, stage guides, cheat sheets, and visual assets
author Paula Silva, AI-Native Software Engineer, Americas Global Black Belt at Microsoft
date 2026-04-24
version 1.3.0
status approved
tags
kit
teams
templates
hackathon
datacorp

🚀 Team Repository Kit

The SIFAP Journey: 30 years of legacy, 1 day to modernize

Hackathon DATACORP 2026 — Team Starter Kit

status: approved version: 1.3.0 personas: 25 stages: 4 event: April 28 2026 use: internal

📍 Start here if you are a hackathon participant.

  1. First: follow SETUP.md to create your team's GitHub repo, configure Copilot, and install Spec-Kit + Specky (30 minutes).
  2. Then: read TEAM-FLOW.md to understand how the 10 of you work together.
  3. Finally: open your persona kit in persona-kits/<your-role>/README.md, then go to Stage 1 at 01-arqueologia/GUIDE.md.

Setup Guide · Team Flow · The SIFAP Story · Your Mission · Your Team · Stage Flow


📑 Table of Contents

  1. The SIFAP Story
  2. Why SIFAP Must Change
  3. Your Mission in 8 Hours
  4. Your Team
  5. The Four Stage Journey
  6. The 4 SDLC Agents
  7. From Legacy to Modern
  8. Folder Structure
  9. How to Use This Kit
  10. Related Repositories
  11. References

🎬 The SIFAP Story

Picture this. It is 1995. A small team of engineers at DATACORP builds a system to pay social benefits to millions of citizens. They pick the best tools available at the time: Natural, a language designed for fast data processing, and Adabas, a hierarchical database known for reliability. They name it SIFAP, the Sistema Integrado de Folha de Pagamentos.

The system works. It pays citizens on time, month after month. The people who built it move on. New people are hired. The business changes. Laws change. Payments rules change. Every few years, someone patches SIFAP to handle a new rule, a new benefit category, a new exception. Nobody rewrites it. Nobody dares to. It just keeps running.

Fast forward to 2026. SIFAP is still running. It processes millions of payments every month. But it is not the quiet hero it used to be. It has become a burden.

  • The original architects are long gone.
  • The source code has abbreviations nobody remembers the meaning of.
  • Every change requires a week of testing, because there are no automated tests.
  • The mainframe licenses cost the company more every year.
  • Citizens want to check their benefits from a phone. SIFAP has no API.
  • Partners want to integrate. SIFAP has no modern endpoint.
  • Operators still use green screen terminals. Nobody under 35 knows how to navigate them.

That is where your team comes in. You have 1 day, 100 developers, 10 teams, and a box full of AI tools. Your mission is to prove that legacy modernization, the kind that used to take 2 years and a consulting army, can now happen in hours, with the right approach.

Welcome to the Hackathon DATACORP 2026.


⚠️ Why SIFAP Must Change

5 pain points of the legacy SIFAP system

# Pain Point Business Impact
1️⃣ Knowledge silos A handful of people know how SIFAP works. When they leave, the company is exposed.
2️⃣ Cryptic code 8 character abbreviations, no comments, no docs. Every change starts with archaeology.
3️⃣ Zero tests Changes are gambles. A fix in one module silently breaks another from 15 years ago.
4️⃣ Costly licenses Mainframe licenses consume budget that should fund innovation.
5️⃣ No integrations Mobile, partners, modern channels. All blocked by the monolith.

💡 Analogy. Legacy SIFAP is like a 30 year old family car that still runs but needs a full tank of gas, a carburetor adjustment every month, and only one mechanic in town can service it. Modernizing is not about buying a faster car. It is about trading it for a reliable electric vehicle the whole family can drive.


🎯 Your Mission in 8 Hours

Your team has one clear goal. Rebuild SIFAP with a modern stack, prove that AI assisted Spec Driven Development delivers, and ship something demo worthy by the end of the day.

Deliverable Format Stage
🗺️ Legacy map and glossary Markdown in 01-arqueologia/ Stage 1
📜 Modern specification EARS requirements + ADRs in the reference modern spec (provided by facilitators) Stage 2
💻 Working prototype Java 21 + Next.js 15 code Stage 3
☁️ Infrastructure and CI/CD Terraform + GitHub Actions Stage 4
🎤 Final pitch 5 minute demo End of day

👥 Your Team

25 persona kits available for your team

Each team has 10 people. The facilitators provide 25 persona kits. You can pick any combination that fits your team and the work you need to do.

🌟 Recommended Core (5 personas, 2 people each)

Persona Role Primary Tool Delivers
🧠 Tech Lead Architecture and decisions Specky SDD + Copilot Chat ADRs, reviews, alignment
Backend Developer Services and APIs Java 21 + Spring Boot REST API, domain, services
⚛️ Frontend Developer Web experience Next.js 15 + TypeScript UI, pages, API integration
🛠️ DevOps Engineer Infra and pipelines Terraform + GitHub Actions IaC modules, CI/CD
🧪 QA Engineer Quality and coverage Vitest + pytest Test suite, quality gates

🎒 Need more specialized roles? The persona-kits/ folder has 25 kits including Product Owner, UX Designer, DBA, SRE, AppSec Engineer, Compliance Auditor, and more. Pick what your team needs.


🗺️ The Four Stage Journey

4 stage flow from archaeology to evolution

flowchart LR
    S1["🔍 Stage 1<br/>Archaeology<br/>3h"]
    S2["📜 Stage 2<br/>Modern Spec<br/>2h"]
    S3["💻 Stage 3<br/>Implementation<br/>2h"]
    S4["🚀 Stage 4<br/>Evolution<br/>1h"]
    S1 --> S2 --> S3 --> S4
    S4 --> D["🎤 Demo<br/>and Pitch"]
    classDef stage fill:#0f172a,stroke:#334155,color:#e2e8f0
    class S1,S2,S3,S4,D stage
Loading
Stage Folder What your team does
🔍 Archaeology 01-arqueologia/ Explore SIFAP, map rules, find the hidden mysteries
📜 Modern Spec 02-spec-moderna/ Write EARS requirements, capture ADRs, design the target
💻 Implementation 03-implementacao/ Build Java backend, Next.js frontend, PostgreSQL schema
🚀 Evolution 04-evolucao/ Terraform to Azure, GitHub Actions pipeline, final demo

🤖 The 4 SDLC Agents

If the persona-kits are the vertical axis — one kit per role — the 4 stage agents are the horizontal axis — one agent per stage of the day. Every persona uses every agent, but the spotlight shifts as the clock advances. Agents know how to modernize Natural/Adabas systems; they do not know anything specific about your legacy system. That knowledge emerges from your investigation.

Agent Stage Mission Kit
@archaeologist-agent 1 — Archaeology Read legacy code, extract rules, map dependencies, catalog mysteries agent-kits/01-archaeologist/
@architect-agent 2 — Modern Spec Carve bounded contexts, write EARS, generate ADRs, design Modular Monolith agent-kits/02-architect/
@builder-agent 3 — Implementation Translate legacy to Java 21, generate JPA, write tests, build REST + Next.js agent-kits/03-builder/
@evolution-agent 4 — Evolution Write GitHub Issues for Copilot Agent, review AI PRs, set up CI/CD and IaC agent-kits/04-evolution/

See docs/persona-agent-matrix.md for the full 10×4 mapping of who uses what.

See docs/4-agents-explained.md for the anatomy of each agent.

See docs/sdlc-flow-guide.md for the complete integration map — who does what, with which agent, at which stage, with every handoff and deliverable in sequence.


🏗️ From Legacy to Modern

Legacy to modern architecture transformation

💡 Think of it as a kitchen renovation. The legacy SIFAP is a kitchen from 1995: functional, but tight, hard to clean, with appliances only one brand still services. The modern SIFAP is an open, modular kitchen where every appliance speaks a standard protocol, and anyone can plug in new devices.


📂 Folder Structure

📁 Path 🎯 Purpose
SETUP.md 🚀 Read first. Step-by-step: create repo, configure Copilot, install Spec-Kit + Specky
TEAM-FLOW.md 🤝 Daily timeline, handoffs, the 20-minute rule
personas/ 👤 The 10 persona cards — read your role
01-arqueologia/ 🔍 Stage 1: legacy code archaeology guides and Copilot prompts
02-spec-moderna/ 📜 Stage 2: EARS specification templates and ADR scaffolds
03-implementacao/ 💻 Stage 3: implementation scaffolding and Copilot Agent prompts
04-evolucao/ 🚀 Stage 4: Terraform guides and CI/CD templates
cheat-sheets/ ⚡ Quick reference cards: Copilot modes, Specky workflow, model routing
persona-kits/ 🤖 25 Copilot agent kits, each with agents, prompts, skills, and a README
agent-kits/ 🤖 4 SDLC stage agents (archaeologist, architect, builder, evolution)
plugins/ 🔌 Shared Copilot plugins (GitHub Issues, Azure Boards sync)
scripts/ 🛠️ setup.sh (bootstrap) and check.sh (run all CI gates locally)
docs/ 📚 Team-curated docs: ADRs, glossary, runbook, agent guides
assets/ 🖼️ SVG illustrations used across this kit
.github/ ⚙️ Issue templates, PR template, Copilot instructions, workflows
.devcontainer/ 📦 Pre-configured dev environment (Java 21, Node 20, Docker)

🛠️ How to Use This Kit

📦 Reference materials provided by facilitators

The stage guides mention these external folders. Facilitators distribute them at the start of the event, or you run ./scripts/setup.sh to clone them as symlinks. Either way, the paths in the guides will work.

Stage Folder Purpose
1 legacy/ Natural programs, Adabas DDMs, legacy docs (sifap-legacy)
2 Reference modern specification (read on GitHub: reference modern spec (shared by facilitators))
3 prototype/ Reference prototype (Java + Next.js)
4 infra/ Reference Terraform modules for Azure

🚀 Getting started

Full step-by-step setup is in SETUP.md — covers GitHub repo creation, Copilot activation, Spec-Kit + Specky installation, branch strategy, and the 13-item smoke test.

Quick summary (full instructions in SETUP.md):

# 1. Create your team's GitHub repo — empty, under YOUR account (one person, once)
gh repo create hackathon-team-XX --private

# 2. Bootstrap from this kit
git clone https://github.com/paulasilvatech/hackathon-datacorp-team-kit.git kit
cp -R kit hackathon-team-XX
cd hackathon-team-XX
rm -rf .git
git init -b main
git remote add origin https://github.com/<YOUR_GITHUB_USER>/hackathon-team-XX.git

# 3. Bootstrap script (clones sifap-legacy, sets up symlinks)
./scripts/setup.sh

# 4. First commit and push
git add -A && git commit -m "chore: bootstrap team kit" && git push -u origin main

# 5. Open in VS Code with the dev container
code .
# Then: Ctrl+Shift+P > "Dev Containers: Reopen in Container"

⚠️ Heads up. The reference prototype and Terraform modules are delivered by facilitators at the start of Stages 3 and 4 respectively. Do not attempt to fetch them from this repository.

🎯 Recommended reading order

  1. 🚀 SETUP.md — get your environment ready (30 min)
  2. 🤝 TEAM-FLOW.md — understand how the 10 of you collaborate
  3. 📖 Your persona card in personas/<your-role>.md
  4. 🤖 Your Copilot agent kit in persona-kits/<your-role>/README.md
  5. ⚡ Cheat sheets in cheat-sheets/ (keep them open in a tab)
  6. 🔍 Stage 1 guide in 01-arqueologia/GUIDE.md

🔗 Related Repositories

Repository Purpose
hackathon-datacorp-team-kit 🎒 This repository — team starter kit
sifap-legacy 🗄️ Read-only legacy SIFAP code (Natural + Adabas) for Stage 1

📚 References

Reference materials live in the main event repository and are shared by facilitators.

📖 Resource 🎯 What you get
Hackathon Blueprint Event plan, schedule, rubric
SIFAP Modern Specification Complete EARS spec for the target system
Reference Prototype Working Java + Next.js implementation
Facilitator Playbook Session scripts, coaching cues, fallback plans

Navigation

Previous Home Next
Team Flow This kit Stage 1: Archaeology

Authored by Paula Silva, AI-Native Software Engineer, Americas Global Black Belt at Microsoft. Version 1.3.0, April 2026.

About

Team starter kit for Hackathon DATACORP 2026: personas, skills, agents, plugins, cheat sheets, and 4-stage workflow (Arqueologia, Spec Moderna, Implementação, Evolução).

Topics

Resources

Stars

Watchers

Forks

Contributors