Merit Flow: The Culture & Ethical Orchestrator 🚀

Hackathon Entry: IBM Dev Day AI Demystified 2026 Theme: AI Demystified - From Idea to Deployment Team Team-Semantic-Snipers-IBM-Dev-Day-AI-Hackathon-2026

🎯 The Vision

Merit Flow is an agentic AI solution that fosters a transparent, healthy, and productive work environment. Instead of using AI for surveillance, we use it for enablement.

We bridge the gap between "doing work" and "being recognized" by orchestrating the flow of merit across the organization.

⚠️ The Problem

• The Recognition Gap: Hard work often goes unnoticed in remote/hybrid setups.
• Stagnant Growth: Employees have access to course libraries (Coursera/edX) but lack the context on what to learn and when.
• Delayed Feedback: Annual surveys are too slow to catch burnout or toxic culture trends.

💡 The Solution

Merit Flow operates as an intelligent layer inside IBM watsonx Orchestrate, connecting your communication tools (Slack/Teams) with HR systems and Learning Platforms.

1. Recognition Engine 🏆

• What it does: Automates "Kudos" and rewards.
• How: When a project is marked "Complete" in the system, the Agent triggers a workflow to prompt the manager or peers for recognition, logging it directly to the employee's merit profile.

2. Growth Pathfinder 🧭

• What it does: Context-aware education recommendations.
• How: The Agent analyzes recent project logs. If an employee struggles with a specific task (e.g., "API Integration"), the Agent queries watsonx.ai to recommend a specific, bite-sized course module to bridge that gap immediately.

3. Culture Pulse ❤️

• What it does: Ethical sentiment tracking.
• How: Instead of spying on messages, the Agent conducts anonymous "Pulse Checks." It uses Granite models to aggregate sentiment data and report "Culture Health" to leadership without exposing individual identities.

🛠️ Tech Stack

• Core Platform: IBM watsonx Orchestrate - Managing the agent workflow and tool connections.
• Intelligence: IBM watsonx.ai - Powering the sentiment analysis and recommendation logic using Granite-3-8b-instruct.
• Integration: Custom OpenAPI skills connecting to a mock HR Data backend.

🚀 Getting Started (Dev)

To run this project locally for development:

1. Clone the repo:

   git clone [https://github.com/JayR1031/merit-flow.git](https://github.com/JayR1031/merit-flow.git)

2. Mock Data Setup: Load the data/mock_employees.csv into your local testing environment.
3. API Spec: Import openapi/merit-flow-api.yaml into your watsonx Orchestrate "Custom Skills" panel.

👥 The Team

• Architect: @JayR1031
• Agent Engineer & Frontend: @Gustavo-Rubio
• AI & Prompts: @shivansh052k
• Tools, Integration & Data: @Sadakakarla

Demo Link:

https://youtu.be/86QonsRbEw4

---

Built with ❤️ for the IBM Dev Day Hackathon 2026.