Building a Knowledge-Sharing Platform for Teams

Everyone on our team has their own Claude. But they all share the same memory. Most teams use AI the same way: someone opens a chat, explains the context from scratch, gets a response, closes the chat. Next day, same thing. Different person, same explanation. The AI never learns, and every conversation starts at zero. At Expanly, we built a shared Knowledge Base that every team member's AI reads before answering anything. It contains our positioning, product details, customer cases, sales arguments, brand voice, even how we talk about competitors. When one of us prepares for a sales meeting, their Claude already knows the relevant customer references and what worked / what to avoid in similar conversations before. When a new team member joins, their Claude can answer questions about our product, customers, and processes from day one. The difference isn't speed. It's consistency. Five people using AI with a shared knowledge layer produce work that feels like it came from the same company. Without it, you get five different versions of the truth. The best part: it's alive. When someone learns something new from a customer call or a sales meeting, it goes into the Knowledge Base. Next time anyone's AI touches that topic, it already knows. If you want to try this with your team, here's how to start: 1. Create a shared Git repository with your key documents: positioning, product info, customer details, sales playbook, anything you'd explain to a new hire. 2. Add an instruction file that tells the AI what your company does, how it should communicate, and where to find context. In Claude, this is typically a CLAUDE.md file. In ChatGPT, an AGENTS.md. Think of it as onboarding for your AI. 3. Connect each team member's AI assistant to read from that same repository. Everyone gets the same foundation, but can use it for their own work. 4. Keep it alive through pull requests. When someone learns something new from a customer call or a sales meeting, they propose an update. The team reviews it, merges it, and every AI in the company knows it from that moment on. 5. Review it monthly. Remove what's outdated, add what's missing. A living knowledge base beats a perfect document that nobody updates. The compound effect is real. Every week, every AI in the company gets a little smarter because the shared knowledge grows. Not just faster work, but more systematic work across the entire team.

Most companies make the same expensive mistake with knowledge. They split it across different systems: - Confluence for employees - Zendesk for customers - Partner portals for resellers - Bespoke AI initiatives to satisfy execs Every time you duplicate or silo knowledge, you: - Increase maintenance costs (multiple teams re-creating the same content). - Erode trust (different audiences get different answers). - Block AI effectiveness (fragmented data = hallucinations + bad retrieval). - Slow down enablement (new hires, partners, and customers all relearn the same basics in different places). 💡 The smarter path: Build once, reuse everywhere. A single knowledge platform can underpin: - Employees → SOPs, onboarding, troubleshooting - Partners → playbooks, certifications, product updates - Customers → help centers, product docs, self-service flows - AI → shared foundation that powers both conversational agents and search The result of unification isn’t just efficiency—it’s scalable enablement: - A support answer written once trains reps, partners, and customers simultaneously. - AI agents pull from a single trusted source instead of guessing. - Every interaction (employee Q, partner escalation, customer ticket) feeds back into the same foundation, compounding value. Companies that unify knowledge don’t just save costs—they create a system where knowledge and content create scalable growth. #knowledgemanagement #customerenablement #employeeenablement #partnerenablement #customerservice

Internal data marketplaces promise to tear down silos and unlock value by letting teams share data freely. It’s a compelling vision: seamless collaboration, endless innovation. But the reality often falls short. The problem? Misaligned incentives. Teams understand their own data very well. But making it usable for others? That’s a heavy lift. Cleaning, documenting, and standardizing data takes time, effort, and resources. What’s in it for the providers? Usually, not much. The result is a one-way street: everyone wants clean and usable data, but no one wants to put in the effort to provide it. Organizations typically try two approaches to address this mismatch, but neither works: ▪ “Share by default” rules: Although universal access makes data technically available, it doesn't make it usable. Without proper cleaning and documentation, teams are left with a sea of unusable data. Compliance happens on paper, but collaboration remains elusive. ▪ Artificial incentives: Internal credits or transfer prices sound clever, but often backfire. They create bureaucracy, invite gaming the system, and rarely inspire genuine engagement. If the logic of “share for everyone’s benefit” doesn’t work, the solution lies in aligning data sharing with personal incentives. Leaders can make a real difference here: ▪ Tie data sharing to success: Make collaboration a key factor in performance reviews, promotions, and team goals. Aligning sharing with personal and team success makes participation natural. ▪ Make it easy: Provide tools and training to lower the effort required for cleaning and documenting data. ▪ Lead by example: Leaders should model openness, sharing their own data and championing collaboration. ▪ Celebrate contributions: Recognize and reward teams that contribute meaningfully. Visible recognition builds momentum. ▪ Cut the red tape: Simplify policies and processes. When sharing is easy, adoption follows. The bottom line: Building a successful data marketplace isn’t about enforcing rules or adding complexity. It’s about leadership creating the right incentives by simplifying processes, and making sharing an opportunity rather than an obligation. When data marketplaces fail, it's not because the concept is flawed. It's because leaders didn’t step in to make them work.

Every time you ask your AI a question, it forgets everything it learned the last time you asked. That's how most AI knowledge systems work today. They're called RAG systems — Retrieval Augmented Generation. You upload documents. The AI searches fragments on every query. Builds an answer from scratch. Closes the session. Next question? Same process. Zero accumulation. Andrej Karpathy — the person who built Tesla's AI vision system and co-founded OpenAI — published something quietly last week that might change this fundamentally (https://lnkd.in/dNB5swS8). He calls it "LLM Wiki." The concept: Instead of searching your documents every time, the LLM builds and maintains a persistent knowledge base. A structured wiki of markdown files with cross-references, concept pages, and contradiction flags. New document arrives? The LLM doesn't just index it. It reads it, extracts the key information, updates existing pages, notes where new data contradicts old claims, and strengthens the evolving synthesis. His key insight: "The tedious part of maintaining a knowledge base is not the reading or the thinking — it's the bookkeeping." Cross-references. Keeping summaries current. Noting when new data overrides old claims. That's exactly what nobody in your team has time for. And exactly what LLMs don't get bored doing. What this looks like in practice: Step 1 — Pick one knowledge domain. Not everything. One area where your team wastes time re-finding information. Customer onboarding. Product specs. Compliance requirements. Step 2 — Set up the structure. Claude Projects or Obsidian + Claude Code as the wiki layer. Raw sources go in one folder. The LLM-maintained wiki lives in another. Step 3 — Feed sources one at a time. Let the LLM summarize, cross-reference, and file. You review. Redirect. Ask follow-up questions. The wiki grows with every session. Step 4 — Query against the wiki, not the raw documents. Answers are faster, more contextual, and cite specific pages. The open-source project Graphify already implements this pattern — claiming 70x fewer tokens needed to answer questions compared to raw-folder RAG. For now, this is still a developer concept. No plug-and-play enterprise solution exists yet. The benchmarks are anecdotal, not peer-reviewed. The wiki could drift or hallucinate if not curated. But the direction is clear: AI that builds knowledge, not AI that searches it. Save this for when your team's knowledge base fails you again. — 📌 Save this post for later ♻️ Share it to inspire your network Follow Hartmut Hübner, PhD for AI insights that work.

I believe you should place Nex’perts at the heart of your knowledge strategy. It’s tempting to build your knowledge management program around your experts. After all, they’re the ones with the deepest experience. So why not just ask them to contribute content to your intranet or manage sections of your knowledge management platform? And if you can get their time, that’s great. But the reality in most firms is that your experts are also your most billable people. They’re in high demand with clients and project teams. Their calendars are full. And while they may want to share what they know, they usually don’t have the time to do it consistently—or the bandwidth to structure and maintain a knowledge base over time. It’s also not a long-term solution. Relying on experts doesn’t create pathways for knowledge transfer. It doesn’t prepare your next generation of leaders. It just centralizes knowledge in the heads of a few already overloaded people. That’s why we place Nex’perts at the center of the process. A Nex’pert isn’t the current practice leader—they’re the one who’s next. Someone with around 10 years of experience. Not new, but not yet senior. They’ve done the work, they’ve got context, and they’re motivated to grow. They’re also close enough to day-to-day project work to understand what’s useful and what’s not. We ask each Nex’pert to take ownership of a domain—say, Sustainability. They spend time with the expert, ask questions, and pull that knowledge out of their head. They organize it, maintain it, and connect with others across the firm doing similar work. Over time, they become the go-to person in that space—not just for information, but for relationships and context. And when the expert retires or moves on, the Nex’pert is already there, ready to step in. You haven’t just preserved knowledge—you’ve grown a leader. The Nex’pert model makes knowledge sharing sustainable. It keeps your content fresh, your people engaged, and your leadership pipeline strong. And it turns your knowledge strategy into a developmental strategy, too. We unpack this model more in my conversation with Evan Troxel on the TRXL Podcast—watch the full episode if you want to go deeper: 💡👉 https://lnkd.in/g2EdftND Thank you to Carla O'Dell of APQC for introducing us to this idea! #AEC #KnowledgeManagement #Intranets #SmarterByDesign

80% of GenAI projects are failing. Ours was a success… Why? We found that making GenAI a success is about: - more than implementing a technology, - addressing real needs, - having the right team, - a thoughtful plan. Personal story: While working with a reduced workload at KSW, I often spent too much time searching for answers on topics like: - insurance coverage, - pension plans, and - vacation days. Here’s how Clara made a difference: ✅ Instant Answers: Clara quickly responded to HR questions, making it easy to find what I needed. ✅ Direct Access: She guided me to the proper documents, saving precious time. ✅ Improved Experience: Far better than previous bots that often frustrated me. What made this project work? We focused on three key steps: 1️⃣ Involve Stakeholders Early Initially, We partnered with HR to ensure Clara met real employee needs. 2️⃣ Start Small, Think Big Today, Clara answers HR-related questions. Tomorrow, she’ll cover all documented knowledge. 3️⃣ Build a Dedicated Team A motivated, passionate team was essential to making her effective. Results? - Faster access to information, - higher-quality answers, and - increased employee satisfaction. Finding the correct information is no longer frustrating, making work easier for everyone. Success isn’t just about the technology – it’s about solving real pain points. P.S. Does your organization use GenAI to make knowledge sharing easier? Let’s share experiences!

I never answer team member Slack DMs.   Instead, I ask them to repost it in a public channel.   I can practically hear the confusion through the screen. Did the CEO just...redirect me? Yes, I did, and for a very specific reason.   Questions, decisions, updates, discussions...everything happens on open channels where others can see it. This is an unwritten rule at Smile.io that throws new hires for a loop every single time.   When you work in an office, you overhear conversations and learn without even trying. Someone chatting over coffee, a teammate sharing a win, a question shouted across the desk. You pick up context just by existing in the same space.   But when you’re working remotely, all that knowledge gets trapped in DMs and 1:1 calls.   The sales team learns something important about customer behavior, but product never hears about it. Engineering discovers a workaround, but support is still manually fixing the same issue.   You only hear what you’re explicitly told, and that’s dangerous for team alignment, context, and growth.   That’s why, at Smile, we live by one core value: If it doesn’t have to be private, say it in public.   As a result, we’ve seen:   • Faster onboarding: New hires don’t need hours of training sessions; they ramp faster by observing real conversations as they happen. • Shared context: Teams better understand each other’s workflows, roadblocks, and the ripple effects of their decisions. • Better questions: When you know 50 people might see your question, you think before you type. • Searchable knowledge: Everything becomes documentation. That debugging session from six months ago? It’s right there in the thread, open to all.   When information lives in DMs, you're building a company where only half the team knows what's happening.

Guru's Knowledge Agents just got an exciting new capability: knowledge sharing.   I've spent 27 years building enterprise software. The hardest problem used to be connecting systems—now it's making sure what flows between them is actually accurate.   That's exactly what this new capability is built around.   Knowledge Agents have always connected and surfaced your company's knowledge. Now they can deliver it outward—pushing verified, continuously improving knowledge directly into the other systems your teams and AI tools rely on every day.   And everything that flows out through this has been through the loop: corrected by your experts, deduped, checked for freshness, permissioned.   Here's what that looks like in practice:   1. Help Centers   Customer-facing content published directly from your source of truth. When something changes internally (a pricing update, a policy revision, a product release), the help center reflects it automatically, because the knowledge flowing there is already verified and current.   2. CRM Systems   Deal context, customer status, and account intelligence stay current because they're fed by verified knowledge rather than manual data entry from overloaded reps. Your sales team's AI copilot and your support team's agent are both pulling from the same source of truth.   3. Work Management   Tasks in Asana, Jira, and similar tools are created with full context (the relevant background, the latest decisions, the right links) instead of a one-line title that forces someone to go hunting.   4. Slack & Teams   Verified context shows up in the right channels proactively, without anyone having to search for it. The system knows the team needs it.   Our Knowledge Agents deliver accurate information to both human-facing and agentic tools. Your support bot, your sales copilot, and your engineering assistant all draw from one verified, continuously improving knowledge base.   Guru's Knowledge Agents just got an exciting new capability: knowledge sharing (More info --> https://lnkd.in/erK7RVEn ) 

🧠 5 Knowledge Management Pain Points — and How Whelp AI Solves Them In large organizations, knowledge is everywhere, but finding the right piece at the right time: that’s the real challenge. From buried documents to siloed systems, enterprises lose time, clarity, and momentum every day. Here are the top 5 pain points in enterprise knowledge management — and how Whelp AI turns each one into a strategic advantage. 1. 🔍 Scattered Information Across Tools The Problem: Documents in Google Drive, policies in Notion, conversations in Slack, spreadsheets in Excel — and no single place to search across them all. How Whelp AI Helps: Whelp connects to your existing tools and creates a unified, conversational interface. Employees can ask questions like: “What’s our Q3 pricing strategy?” “Show me the onboarding checklist for new hires.” And Whelp pulls the answer from wherever it lives, instantly. 2. 🧱 Knowledge Silos Between Teams The Problem: HR has one system, Finance another, Legal a third. Valuable insights stay locked inside departments, slowing collaboration and decision-making. How Whelp AI Helps: Whelp builds a knowledge graph that links concepts, documents, and decisions across teams. It breaks down silos by making institutional knowledge searchable and shareable, without changing how teams work. 3. 🕰️ Time Lost Searching for Answers The Problem: Employees spend hours each week hunting for information — asking colleagues, digging through folders, or recreating work that already exists. How Whelp AI Helps: Whelp turns search into conversation. Instead of keywords, employees ask questions in plain language and get contextual answers. It’s like having a smart teammate who knows everything your company knows. 4. 🧓 Tacit Knowledge Walks Out the Door The Problem: When experienced employees leave, their insights often leave with them. Tacit knowledge — the stuff that’s never written down — is hard to capture. How Whelp AI Helps: Whelp surfaces insights from conversations, decisions, and documents over time. It builds a living memory of your organization, so knowledge isn’t lost — it’s preserved, searchable, and reusable. 5. 📉 Low Engagement with Knowledge Systems The Problem: Traditional knowledge bases are clunky, hard to navigate, and rarely updated. Employees don’t use them — and don’t trust them. How Whelp AI Helps: Whelp is built for natural interaction. It feels like chatting with a colleague, not querying a database. And because it’s integrated into tools teams already use (like Slack and Notion), adoption is frictionless. 🚀 The Bottom Line Whelp AI doesn’t just organize your knowledge — it activates it. By turning fragmented data into intelligent dialogue, Whelp helps every employee move faster, make smarter decisions, and stay aligned. Ready to chat with your data? Whelp AI makes enterprise knowledge searchable, conversational, and always available.

In this video, you’ll learn how to build and use an AI-powered knowledge system using MCP Server, Confluence, AWS Bedrock, and Cursor. We’ll start by setting up the MCP server from scratch — including generating API tokens, configuring access, and connecting Confluence as a data source. Then, I’ll show you how to integrate MCP with Cursor so you can query your documentation directly from your editor. Next, we’ll explore why enterprises need a Knowledge Base — especially when documentation is spread across multiple Confluence spaces, RCA guides, runbooks, and acquired company data. You’ll learn how to: Create a Knowledge Base in AWS Bedrock (Build section) Use the Confluence connector to ingest enterprise documentation Store embeddings in an OpenSearch vector store Test retrieval using the Bedrock Chat UI Finally, we’ll connect everything together — using MCP to interact with AWS Bedrock and query your Knowledge Base directly from Cursor. By the end of this video, you’ll understand how to: • Turn scattered documentation into a semantic knowledge system • Ask natural language questions over internal docs • Retrieve runbooks, RCA guides, and architecture patterns instantly • Build an enterprise-ready AI agent workflow This is especially useful for data teams, platform teams, and organizations managing large-scale documentation. If you’re working with AI agents, RAG systems, or enterprise knowledge platforms — this video is for you.