Data Neighbor Podcast

• Ep18: Open-Source LLMs vs. ChatGPT: Which One Should You Use?

  AI is evolving faster than ever—and open-source AI models are catching up to proprietary models at an incredible pace. In this episode of the Data Neighbor Podcast, we sit down with Maarten Grootendorst, co-author of Hands-On Large Language Models with Jay Alammar, DeepLearning.AI instructor, and creator of BERTopic and KeyBERT, to break down the real differences between open-source and closed-source AI models.We’ll discuss how LLMs (Large Language Models) evolved from bag-of-words and Word2Vec to modern transformer-based models like BERT, GPT-4, DeepSeek, LLaMA 2, and Mixtral. More importantly, we explore when open-source AI models might actually be better than proprietary models from OpenAI, Google DeepMind, and Anthropic.Hands-On Large Language Models (Maarten’s Book): https://www.amazon.com/Hands-Large-Language-Models-Understanding/dp/1098150961DeepLearning.AI Course: How Transformer LLMs Work: https://www.deeplearning.ai/short-courses/how-transformer-llms-work/Maarten’s AI Newsletter: https://newsletter.maartengrootendorst.com/Connect with us!Maarten Grootendorst: https://www.linkedin.com/in/mgrootendorst/Hai Guan: https://www.linkedin.com/in/hai-guan-6b58a7a/Sravya Madipalli: https://www.linkedin.com/in/sravyamadipalli/Shane Butler: https://www.linkedin.com/in/shaneausleybutler/

  --------  

  1:00:15
• Ep17: How to Use AI and Build Data Science Teams

  In this episode of the Data Neighbor Podcast, we sit down with Justin Chen, Senior Director of Growth Data Science and former Head of Data Science & Engineering at Coinbase, to dive deep into how to build scalable, high-impact data teams. We get into the mind of a data leader who has built numerous reputable data organizations to understand principles, lessons, and challenges every company faces. Justin shares invaluable insights on:- Go slow to go fast – Why early-stage speed can create long-term inefficiencies.- The evolution of data science – How different orgs (growth, core, platform) function within companies.- The myth of the unicorn data scientist – Why hiring for specialization is key.- AI's impact on data science – How automation and AI tools are shaping the future of analytics.- Getting a seat at the table – How data professionals can move from support roles to strategic leadership.Justin also shares his firsthand experience of building an AI-powered data copilot at Coinbase to streamline analytics workflows, offering a sneak peek into how AI will shape the next generation of data teams. Whether you’re a data scientist, engineer, or aspiring leader, this conversation is packed with practical advice and industry wisdom you won’t want to miss!Connect with Justin: https://www.linkedin.com/in/mingc/Connect with Hai, Sravya, and Shane (let us know which platform sent you!):Hai: https://linkedin.openinapp.co/4qi1rSravya: https://linkedin.openinapp.co/9be8cShane: https://linkedin.openinapp.co/b02fe#DataScience #AI #aiengineering #TechLeadership #MachineLearning #GrowthDataScience #CareerAdvice #Analytics #Hiring #DataEngineering #DataDriven #DataTeams #TechPodcast #aiagents

  --------  

  53:33
• Ep16: AI is Breaking the Internet - for Better or Worse

  AI is changing everything—including how we moderate content online. In this episode of the Data Neighbor Podcast, we sit down with Sugin Lou, a Staff Data Scientist at Cash App and former Nextdoor AI trust & safety expert, to discuss the challenges of AI content moderation, misinformation, and trust & safety in the era of LLMs.If you care about AI trust, AI policy, risk of AI, and AI governance, this episode is for you.More about this episode:What is content moderation? How does AI impact trust & safety? From Facebook moderation to moderation bots and comment moderation, companies rely on AI-powered moderation tools to detect AI-generated content, deepfakes, misinformation, and harmful speech. But is AI moderation really working, or is it just scaling misinformation at an unprecedented rate?AI Misinformation & Risk Management:With AI-generated content, fake AI identities, and deepfakes spreading faster than ever, AI-powered disinformation is becoming a serious issue. We explore how AI risk management, AI governance, and AI regulation are trying to catch up before AI trust is lost forever.Trust & Safety in AI:How do platforms like Facebook, YouTube, and Nextdoor determine what content gets removed? How does the moderation process work? And what are the hidden risks of AI trust & safety failures?Evaluating AI Models for Trust & Safety:How do companies evaluate LLMs and ensure AI-generated content isn’t spreading misinformation? We discuss the latest in AI safety, LLM evaluation, and how companies like OpenAI, Google, and Anthropic are handling AI fraud, AI accountability, and AI disinformation.Key Topics Covered:-What is content moderation? AI’s role in trust & safety-AI moderation bots vs. human moderation-Facebook moderation & the future of AI content filtering-The hidden risks of AI-generated content & deepfakes-How AI is breaking the internet—for better or worse-AI misinformation detection & AI disinformation at scale-AI fraud, risk assessment, and AI accountability-How AI safety teams are responding to AI threats-With AI moderation tools, chat moderation, and content filtering AI, tech companies are trying to prevent AI-powered misinformation while balancing --AI ethics, AI regulation, and free speech. But can AI content moderation actually keep up?Connect with us!Sugin Lou: https://www.linkedin.com/in/sugin-lou/ Hai Guan: https://www.linkedin.com/in/hai-guan-6b58a7a/Sravya Madipalli: https://www.linkedin.com/in/sravyamadipalli/Shane Butler: https://www.linkedin.com/in/shaneausleybutler/#AI #ArtificialIntelligence #MachineLearning #AIContentModeration #FacebookModeration #ModerationBot #AITrust #AIAccountability #AIMisinformation #Deepfake #AIRegulation #TrustAndSafety #GenerativeAI #LLMEvaluation #AITrustAndSafety #AICompanions #AIEthics #AIModerationTools #WhatIsContentModeration

  --------  

  57:39
• Ep15: How Top Companies Use Data Science to Build Winning Products

  In this episode of the Data Neighbor Podcast, discover how data science shapes both winning and losing productr. We explore key insights from staff data science expert Anagh Pal with experience at Amazon, Twitter, and Nextdoor. Learn how cross-functional collaboration, data-driven experimentation, and hypothesis testing drive product success and avoid common pitfalls. Whether you're a data scientist, product manager, or just curious about how tech products evolve, this episode reveals essential strategies and lessons.Understanding product analytics and product development is crucial for anyone in data science, product management, or product design. In this episode, we dive into the product roadmap process and how data scientists and product managers collaborate to drive innovation. Learn how data analytics roadmaps influence decision-making, A/B testing, and experimentation to build products that users love. Whether you're wondering "What is data science?", planning your data science roadmap, or looking to advance in product management, this episode provides insights into real-world applications at top tech companies. Discover the key metrics, frameworks, and cross-functional strategies used by leading product managers and data scientists to launch, refine, and scale successful tech products.Connect with Hai, Sravya, Shane, and Anagh (let us know YouTube sent you!):Hai Guan: https://linkedin.openinapp.co/4qi1rSravya Madipalli: https://linkedin.openinapp.co/9be8cShane Butler: https://linkedin.openinapp.co/b02feAnagh Pal: https://www.linkedin.com/in/anaghpal/Learn about:-The role of data science in product success and failure-How data collaboration with PMs and engineers shapes outcomes-Data experimentation, A/B testing, and analyzing key metrics-Earning trust as a data scientist in cross-functional teams-Insights from careers at Amazon, Twitter, and Nextdoor-The future of data science and product development in 2025#datascience #productdevelopment #abtesting #collaboration #techroadmap2025 #dataanalystroadmap #productroadmap #datastrategies #datascientistrole #bigtech

  --------  

  28:32
• Ep14: How DeepSeek Beat OpenAI - The 3 Breakthroughs No One Saw Coming

  AI is evolving faster than ever, and DeepSeek has sent shockwaves through the industry. In this episode of the Data Neighbor Podcast, we sit down with Dr. Raffaele Ciriello, a scholar in information systems at the University of Sydney, to unpack the impact of DeepSeek and its implications for the future of AI innovation, sustainability, and ethics.DeepSeek R1 and DeepSeek V3 have positioned themselves as major competitors to OpenAI's GPT-4, Google's Gemini, and Meta's LLaMA models—but what makes them different? In this episode, we discuss how DeepSeek AI is redefining the large language model (LLM) space, how its open-source AI model challenges proprietary AI, and whether it could be the best free AI alternative. We also explore the financial impact on Nvidia stock, as DeepSeek’s approach proves that cutting-edge AI doesn’t require billions in GPU investment. Plus, we break down how DeepSeek's mixture-of-experts (MoE) architecture is reshaping AI efficiency, its potential risks in AI companionship and generative AI, and what this means for the future of AI regulation and governance.🔗 Links Mentioned:Connect with Raffaele: https://www.linkedin.com/in/raffaele-ciriello/Raffaele’s Google Scholar: https://scholar.google.ch/citations?user=BIJHqJYAAAAJ&hl=enRaffaele’s ResearchGate: https://www.researchgate.net/profile/Raffaele-CirielloLatest article from Raffaele on OpenAI's Deep Research capability: https://theconversation.com/openais-new-deep-research-agent-is-still-just-a-fallible-tool-not-a-human-level-expert-249496Raffaele's article on Compassionate AI Design, Governance, and Use: https://www.techrxiv.org/users/886325/articles/1264666-compassionate-ai-design-governance-and-use🔍 Key Topics Covered:🔥 The rise of DeepSeek and why it’s a game-changer💰 How DeepSeek shattered AI’s cost barriers🆚 Open Source vs. Proprietary AI: What’s at stake?⚡ The environmental impact of AI & the Jevons Paradox🚀 AI regulation, safety, and the road aheadWe also dive into AI companionship, the ethical risks of unchecked AI, and how open-source AI could empower more responsible governance. Don’t miss this eye-opening discussion!Connect with Hai, Sravya, and Shane (let us know which platform sent you!):Hai: https://linkedin.openinapp.co/4qi1rSravya: https://linkedin.openinapp.co/9be8cShane: https://linkedin.openinapp.co/b02feIf you need personalized advice for your data science career roadmap, send us a message on LinkedIn or post a comment below!

  --------  

  45:28

More Technology podcasts

Trending Technology podcasts

About Data Neighbor Podcast