Value Chain Hackers: Integrated 8-Step Plan & VCH-Infra Tools

Christiaan Verhoef

2025-01-20

Welcome to Value Chain Hackers

Empowering Supply Chains Through Collaboration and Innovation

Value Chain Hackers Logo

“Together, we create scalable, sustainable solutions to wicked supply chain challenges.”

About Value Chain Hackers

  • Mission and Vision

    • To bridge the gap between education, industry, and academia by providing a platform that facilitates the development of sustainable and scalable supply chain solutions.
    • To equip students with practical skills and real-world experience in problem-solving, data analysis, and innovative thinking.
    • To empower students and professionals to address complex supply chain challenges by fostering a community of change makers and providing structured frameworks for impact projects.

    Overview of Initiatives

    Who are we?

    • Value Chain Hackers is an innovative incubator dedicated to transforming supply chains through collaborative problem-solving and cutting-edge solutions.

Introduction: Setting the Stage

Purpose of the Lab

  • Mission: Tackle messy, wicked supply chain challenges with solutions that don’t just work—they scale.
  • Vision: Be the bridge between academia, industry, and education to deliver real, measurable impact.

Why It Matters

  • Empowering Change:
    • The tools we use shape how we collaborate, innovate, and grow.
    • Misaligned tools? They’re costly mistakes that derail progress.
  • Collaborative Impact:
    • When we align our tools and processes, trust grows, results multiply, and everyone wins.

Today’s Focus

  1. Overview of the current tools & plan
    • What have we selected, how will we apply it?
  2. A.I and other support tools overview
    • Which Support tools are there and how could we use them.
  3. Shape the Future: Workshops & Training
    • What Training will we be giving them?

Note

“This is about more than tools—it’s about creating a system that fuels our ambitions and delivers the impact we’ve promised.”

Current ++: The Smart Upgrade 🚀

What’s in the Pack?

  • Collaboration: Microsoft Teams, Canva Whiteboard, Canva Prototyping
    • Improved visual brainstorming and structured collaboration.
    • Trough the Tonomy Foundation, we have a Pro-License on Canva with which we can provide the students with their own small team.
    • CANVA IDEATIONBOARD
  • Data Collection: Google Scholar, Excel, OneDrive: And other sources.
    • Optional: See A.I tools
  • Visualization: Canva Pro
    • Elevates presentations with advanced design tools, templates, and AI-driven features.
    • We can give each team their own dedicated team space.
  • Task Management (Optional): Trello, or Focalboard;
    • Lightweight task tracking for teams that are more serious and want more structure.
  • File Sharing: OneDrive
    • Familiar and integrated with Microsoft ecosystem, which the students will know.
    • Optional: Github
  • AI Support: OpenWebUI, Ollama
    • Entry-level AI for brainstorming and lightweight automation.

Key Benefits

  • Problem: Communication is fragmented across email and Teams.
  • Solution: Canva can steamline the whole show.

Why it works

  1. Minimal Disruption: Builds on tools your team already knows and uses.
  2. Quick Wins: Adds impactful tools like Canva Pro with minimal learning curves.
  3. Boosted Creativity: Tools like Canva Pro, Mural, and Excalidraw foster innovation and engagement.
  4. Professional Design Made Easy: Canva Pro allows for high-quality visuals without the need for advanced design skills.
  • Problem: Students don’t know where to get the data

Solution: We give them a lot of sources in advance, We try to use the data that they have collected before

Solution 2 (Advanced) We use A.I tools like Crawl a.i and Seargnx to do data collection.
Note: Due to it being on OneDrive, the data will not be insightfull for A.I tools, or other tools that use a relationalship database. Onedrive is just a folder system after all.

  • Problem: Research is inconsistent and scattered.

Solution: Centralizing in Excel and OneDrive ensures data is organized and accessible.
At a minimum we can make some more data available for them at the start and have a better file structure.

Solution notes: This tool does not fix the research being scattered and will not enable students to very effectively use the research of others, or the data that we have gathered.

  • Problem: Static, uninspiring presentations using PowerPoint.
  • Solution: Canva Pro offers:
    • Magic Resize: Instantly adapts designs for multiple platforms.
    • AI-Powered Tools: Automatically generate text or visual ideas with Magic Write and Dream Lab.
    • Professional Templates: Ensures consistent branding with its Brand Kit.
    • Background Remover: Easily refine images for polished visuals.
    • Shared Storage: Collaborate on designs using unlimited folders and up to 1TB of storage.

Problem: Limited automation and reliance on external AI.

Solution: Focus on making A.I Supportive but not leading

  • Kotaemon (Tested, works): Detects anomalies and generates actionable insights from real-time data for supply chain evaluations.
  • Chainforge(local only, but properly tested!): Automates AI mapping to uncover hidden relationships in workflows and supply chain data.
  • SearxNG(Operational, Fully Functional): Aggregates search results from multiple engines while respecting privacy for diverse research insights.
  • OpenWebUI(Operational, Fully Functional): Local AI assistant for brainstorming, qualitative analysis, and secure conversational support.
  • Flowise:(Tested, Not in production) Simplifies data exploration with a conversational AI interface for intuitive natural language queries.
  • Crawl A.I (Tested, not operational): Automates large-scale web scraping to gather research data quickly and efficiently.

AI and Support Tools Overview

Essential Overview

Current Challenges: Manual workflows, scattered communication, and inefficiencies hinder progress, A.i tooling is found to be unreliable!

  • Opportunity: We use tokens for the Big A.I models to temporary leverage insanely big infrastructure
  • Opportunity: We use many different models and different tools to show the difference between those models, to demonstrate the power of using smaller models
  • Opportunity: We use local hardware, to intergrate the A.I directly into the researcher’s computers;
    • A local deployable A.I set has been developed and is being expanded: It works with n8n, quant, ollama in one

Core AI Tools

  • Kotaemon: Interact with documents seamlessly using Retrieval-Augmented Generation (RAG) technology.

  • ChainForge: Evaluate and compare LLM responses for robust prompt engineering and model selection.

  • Neo4j: Map supply chain networks and visualize relationships.

  • OpenWebUI: Local interface for secure interaction with large language models (LLMs).

  • Flowise: Graphical AI workflow builder for creating intuitive AI applications.

  • Crawl A.I: Automates web scraping to collect large-scale research data.

Tool Purpose Game-Changing Benefit
Kotaemon Document interaction via RAG Makes document exploration intuitive with LLM integrations.
ChainForge Prompt engineering platform Tests and compares prompts across multiple LLMs for robustness.
Neo4j Graph-based visualization Maps supply chain relationships for better scalability.
OpenWebUI Local AI assistant Reduces reliance on external LLMs while ensuring privacy.
BigAgi Workflow automation Automates repetitive processes and accelerates insights.
Ollama Local inference engine Lightweight AI experimentation with high security.
Flowise AI workflow builder Simplifies AI application creation with drag-and-drop nodes.
Crawl A.I Web crawling Collects and organizes large datasets for research purposes.

Tool Purpose How It Assists AI Tools
n8n Workflow automation Automates data flows into and out of AI tools.
Qdrant Vector database Supports embeddings and similarity search for AI pipelines.
Grafana Real-time dashboards Visualizes metrics from AI outputs in actionable formats.
SearxNG Privacy-focused search engine Enhances data collection for AI tools without compromising privacy.

Shape the Future: Workshops & Training

Workshop and Training Overview

  • Create better work, faster by giving students access to these workshops.
  • Focused on learning skills, over doing research.
  • Focused on delivering actionable knowledge for immediate impact in projects.
  • Focused on standalone use-cases due to ecosystem limitations.

Core Training Focus Areas

  1. Prompt Engineering:
    • Crafting effective prompts for OpenWebUI and AI tools.
    • Real-world applications of AI for supply chain problem-solving.
  2. Better Presentations:
    • Leveraging Canva Pro for dynamic, visually engaging presentations.
    • Techniques to integrate storytelling with visuals for stakeholder alignment.
  3. Using OpenWebUI:
    • Hands-on experience to explore local AI capabilities.
    • Learn to use OpenWebUI for brainstorming, FAQs, and secure tasks.

Outcomes

  • Confidence in using A.I tools
  • Practical AI skills without overwhelming participants with new workflows.

Workshop Modules and Formats

Focus Details
Tools OpenWebUI, ChainForge
Objective Learn to design effective AI prompts.
Format Hands-on practice with guided examples.
Outcome Craft precise prompts to unlock AI insights.

Focus Details
Tools OpenWebUI
Objective Explore local AI for secure brainstorming.
Format Interactive session with real-world use cases.
Outcome Understand how to integrate OpenWebUI into workflows.

Focus Details
Tools Canva Pro
Objective Create visually compelling, stakeholder-ready presentations.
Format Guided session with hands-on exercises.
Outcome Produce professional presentations that enhance engagement.

AI and Data Tools

Tool Name Description GitHub Link
Hugging Face Natural language processing and AI model hub. Hugging Face GitHub
LangChain Framework for building applications with language models. LangChain GitHub
Stable Diffusion AI-powered image generation. Stable Diffusion GitHub
Whisper Automatic speech recognition system by OpenAI. Whisper GitHub
Haystack NLP framework for building search systems. Haystack GitHub
FastAPI Web framework for building APIs with Python. FastAPI GitHub
Pandas AI Enhances Pandas with AI capabilities. Pandas AI GitHub
LlamaIndex Tool for integrating knowledge bases with language models. LlamaIndex GitHub
DeepSpeed Optimize and train deep learning models at scale. DeepSpeed GitHub
AutoGPT Autonomous AI agent powered by GPT models. AutoGPT GitHub
Prompt Toolkit Library for building interactive command-line applications. Prompt Toolkit GitHub
Diffusers Library for training and using diffusion models. Diffusers GitHub
BERT Transformer-based language model. BERT GitHub
Transformers Library for working with pre-trained transformer models. Transformers GitHub
Cortex Platform for deploying machine learning models. Cortex GitHub
Dash Framework for building web-based analytical applications. Dash GitHub
Streamlit Create custom machine learning apps quickly. Streamlit GitHub
Ray Framework for scaling AI and machine learning workflows. Ray GitHub
Metaflow Machine learning workflow management tool. Metaflow GitHub
PyCaret Low-code machine learning library. PyCaret GitHub