Danny Thien

The AI Paradox: Hungry for Progress, Stingy with Data 🤖🔐 We all want AI superpowers but are unwilling to feed the beast. Here's the truth: You're excited about AI's potential 🚀 You're using ChatGPT, Midjourney, and more 💬🎨 But you're also hitting that "opt-out" button 🔴 Here's why that's a problem: AI needs data to improve 📊 Opting out limits AI's growth 📉 We risk creating biased, limited AI 🚫 The hard truth: If you want to benefit from AI, you need to contribute. What you can do: Share responsibly 🤝 Engage in the AI conversation 🗣️ Help shape AI's future 🌟 Remember: Ideas without execution are worthless. Don't hoard your data. Be part of the solution. Agree or disagree? Drop your thoughts below 👇 #AI #DataPrivacy #TechFuture #Innovation

6

Time series transformer models in Databricks to get trained on your data : Chronos TimesFM Moirai TimeGPT https://lnkd.in/gh_vwWE2

7

Good overview of some of the terminology and types of problems and how to understand them in #Databricks

1

A useful step by step video on setting up Databricks unity catalog on Microsoft Azure. https://lnkd.in/gUbQqqiP

1

I am researching metrics to monitor the performance of newly implemented Data Governance controls and made a shortlist. As any metric takes time and effort to monitor: which one would you personally focus on? (pick one) 1) Activity-based KPIs (e.g., Data Catalog visits, tickets raised, data requests made, etc.) 2) Coverage-based KPIs (e.g., systems or departments covered, rollout progress, etc.) 3) (Meta-)Data Quality-based KPIs (e.g., timeliness or completeness of documentation) 4) Satisfaction-based KPIs (e.g., NPS, stakeholder feedback, etc.) 5) Business Case-based (e.g., use cases enabled, efforts reduced, etc.)

37

6 Comments

Starting in few minutes

4

During #AutomationDaysAsia2024, we shared our vision and framework for Autonomous Intelligent Automation, highlighting how organizations can transition from intelligence to autonomy. Automation has evolved through four distinct phases: - RPA (Robotic Process Automation): Focused on automating repetitive, rule-based tasks to enhance operational efficiency. - Intelligent Automation (IA): Combined RPA with AI capabilities, such as machine learning, natural language processing, and computer vision, enabling systems to process unstructured data and make decisions. - Agentic Intelligent Automation: Introduced goal-driven agents capable of dynamic workflows and partial autonomy, allowing systems to act and adapt in increasingly complex scenarios. - Autonomous Intelligent Automation (AIA): The ultimate phase, where systems operate "fully" autonomously, continuously learning and optimizing processes without human intervention. At IAC.ai, our journey reflects these phases. We’ve been transforming automation from simple task execution to fully autonomous systems that enable our clients to thrive in today’s unpredictable markets. Our goal is clear: to create adaptive, autonomous workflows that drive value and innovation. Let’s shape the future of work—together. 🚀 📺 If you would like to see my Keynote during the event please check link in comments. #IntelligentAutomation #Automation #AutonomousAgents IAC.ai : the Intelligent Automation Company

25

5 Comments

Another preview of my Discussion with Sanjeev about AI and Data strategy and why Sanjeev loves to talk about AGI 😀 - the full episode will drop this Friday. I can't wait! #bytesofbrilliance #CDOTIMES #datastrategy #metadatalake #CDO

5

2 Comments

What an amazing panel to kick off the PH leg of AIBP’s Conference this year!! Main takeaway: There are 5 key factors for a successful adoption of Gen AI: 1. Data Quality 2. Data Quantity 3. Cost 4. Cultural Change (people’s mindset, training) 5. Data Architecture AIBP ASEAN, AIBP

24

Continuous learning is essential. That’s why I’m excited to share I’ve earned my Inspect Rich Documents with Gemini Multimodality and Multimodal RAG skill badge #GoogleCloudLearning #GoogleCloudSkillBadge

4

Amazon, Google and Azure - Who will dominate the cloud infrastructure as a service market in 2024 and beyond? We conducted a detailed analysis and Amazon Case Study at CDO TIMES looking at how Amazon is succedding on so many levels. This is by design and not by chance. For businesses seeking to replicate Amazon's success, the key takeaways are clear: 👉 Customer Obsession: Always prioritize the customer experience and continually seek ways to exceed expectations. 💰 Value Expansion: Consistently add new features and benefits that align with customer needs and preferences. 🤖 Technological Innovation: Leverage technology to enhance convenience and personalize the customer journey. 🌎 Global Adaptation: Tailor offerings to meet the unique demands of different markets, ensuring relevance and appeal. 📊 Data-Driven Insights: Use data analytics to understand customer behavior and drive strategic decisions. See more in our in depth analysis of strategies and successfactors for services like AWS cloud services and Amazon Prime. If you like this: Like, share, repost and subscribe to our digital insider newsletter on my profile page. Thank you, Carsten I will post the link to the article in the comments. #AWS #Amazon #Googlecloud #Azure #cloudstrategy #digitalstrategy #datastrategy #cdotimes

10

3 Comments

Data Governance Eats AI for Breakfast 🍳 A big thank you to Pierre-Charles Igras for highlighting the insightful Gartner article, “Adopt a Data Governance Approach That Enables Business Outcomes.” 🙏 Check it out: https://lnkd.in/d6MF22wG “Current data governance practices are often too rigid and insensitive to the business context. By 2027, 60% of organizations will fail to realize the anticipated value of their AI use cases due to incohesive data governance frameworks. To avoid this, incorporate best practices into your data governance roadmap and: 📊 🔹 Build an effective governance structure 🔹 Design and deploy effective policies and standards 🔹 Evaluate and improve data governance performance 🔹 Establish a process of iteration and learning” As Head of Data at TotalEnergies EP Danmark (TEPDK), my mission is to fuel a data-driven decision-making culture. This includes not just Data Governance, Data Quality, and Data Architecture, but also Data Literacy, Decision Intelligence, and value creation through timely, well-informed decisions. I believe Data and Analytics should be unified, with joint governance of Data and AI/ML. While Digital Transformation will eventually conclude, Data and Analytics will remain a crucial capability. 🚀 At TEPDK, we’ve just finalized the second revision of our Data Management Plan. This report outlines our new Data Management Operating Model, updates on identified Data Stewards, and defines quality requirements, challenges, and actions for all Business Objects. It also describes our Data Architecture, highlighting the crucial role of our Data Virtualization service in exposing quality-verified “golden data” from 10 different Data Domains to data consumers, including Data Actors, Power BI reports, and both on-premises and cloud applications like our Operations Digital Twin and AI/ML tools. 📑 Recognized as the frontrunner in data culture within the group, TEPDK emphasizes effective data management as crucial for creating reliable and robust GenAI systems. This commitment to data excellence is exemplified by the deployment of the group’s first GenAI-enabled product, developed by the Danish digital team, and by our consistent leadership in the EP Best Innovators Awards for three consecutive years. Ultimately, this drives business value through better-informed decision-making. 🏆 Build a culture of trust to navigate the complex challenges of data, analytics, and AI. Manage risks proactively, adopt adaptive governance, and drive culture change to support data-driven decisions and deliver business value. 🌟 What are your thoughts on the future of data governance? Share your insights in the comments! 💬 #Governance #DataManagement #DataQuality #DataArchitecture #DataLiteracy #DecisionIntelligence #AI #MachineLearning #DigitalTransformation #DataDriven #TotalEnergies #DataAnalytics #BusinessOutcomes #DataCulture #DataStrategy #DataSecurity #DataPrivacy #DataEthics #DataProtection #DataLiberation #DataPolicies

3

Global tech leaders responded to Cloudera on the current state of #EnterpriseAI. Their insights are at once thrilling 😄 and daunting 😲 ✔ Check it out in CDO Magazine https://lnkd.in/gup5dztM

41

1 Comment

Last week, I had the privilege of organizing an intimate data quality meetup in Japan, bringing together 10 data platform managers from enterprises who started their careers as data scientists. The discussions revealed fascinating insights about the reality of data quality in organizations. While many believe aligning with DMBOK (Data Management Body of Knowledge) framework is sufficient for maintaining data quality, our collective experience showed a different story. The real challenge lies deeper - data quality issues are typically symptoms of fundamental organizational pain points. The most critical challenge we identified? Data governance ownership and responsibility. Without clear governance structures, organizations struggle to: 1. Define meaningful data quality criteria 2 .Break free from the endless cycle of data cleaning 3 .Establish sustainable quality management processes Interestingly, while data governance is crucial, individual stakeholders often lack the motivation to participate. A quick Reddit search for "Is data governance boring?" reveals that this sentiment isn't unique to Japanese organizations. This disconnect between recognizing data governance's importance and driving actual engagement presents a fascinating challenge. How does your organization approach this balance? #DataQuality #DataGovernance #DataManagement #AIGovernance

10

Almost 30% of all GenAI projects would stall at POC stage. That's what Gartner says. https://lnkd.in/gjGNdAmx? I wonder why are they attributing this to high upfront cost. There are cheaper architectural approaches like having RAG over a small knowledge base and using graphDB to pre-formulate the relationships between various unstructured datapoints for MVP. Why do people need to fine tune an entire model before getting the sense of ROI? Can anyone enlighten me, what am I missing here?

1

Tune in at 9 AM PT today 🎤 to hear me share my expert insights on data quality, analytics, and AI at #MDQS2024 Here is the link to register👉🏼 https://lnkd.in/gHB2izAf See you soon! 🙏 #DataQuality #AI #DataGovernance

15