Data Neighbor Podcast

Become and Expert in Agentic Analytics and AI Evals at https://dataneighbor.com/Learn more about Index:https://index.app/In this episode of the Data Neighbor Podcast, we sit down with Xavier Pladevall, co-founder of Index, to break down what’s actually changing in business intelligence and what’s not as AI enters the analytics stack.Index is building an AI-powered data platform that blends dashboards, SQL, and an agentic chat experience. But as Xavier explains, the hard parts of data work still show up in the same places: messy data, unclear questions, and getting a room of humans to trust and act on what they see.
In this episode, you’ll learn:- Why most “modern data” talk is disconnected from what teams actually use day to day- Why dashboards remain the default interface for stakeholders- How product design and UI quality change trust in data- What Index 1.0 is solving today, and what Index 2.0 unlocks next- How Index 2.0 makes analytics more proactive with recommended questions- Where AI breaks in analytics and why data quality is still the bottleneck- Why “data slop” compounds AI errors in surprising ways- Why data scientists are not going anywhere, even as AI takes on more tasks- Why SQL is more important than ever, just more abstractedWhat the real future looks like beyond “chat with your data”

#datapodcast #analytics #businessintelligence #datatools #datascience #aiforanalytics #agenticanalytics #productanalytics #dataleadership #dataneighbor #index

Explore Data Neighbor free workshops on AI evaluation and agentic analytics (plus our live, hands on course): https://dataneighbor.com/Learn more about LiveDocs:https://livedocs.com/In this episode of the Data Neighbor Podcast, we sit down with Arsalan Bashir, founder and CEO of LiveDocs, to unpack one of the biggest shifts happening in data right now: tools that don’t just help you analyze data, but start doing parts of the analysis for you.If you care about the future of analytics workflows, this is a must watch.Connect with the team (tell us YouTube sent you!):Shane Butler: https://linkedin.openinapp.co/b02feSravya Madipalli: https://linkedin.openinapp.co/9be8cHai Guan: https://linkedin.openinapp.co/4qi1rConnect with Arsalan:https://www.linkedin.com/in/arslnb/In this episode, you’ll learn:- Why “chat with your data” is the wrong mental model for where analytics is headed- What an “AI data scientist” actually means in practice (and what it doesn’t)- Why context matters more than schema for getting correct results from AI-generated analysis- How LiveDocs blends notebooks, app-building, and agentic workflows in one workspace- What it looks like when AI automates the 0→80% of data work (aka “yak shaving”)- How scheduling, notifications, and lightweight automation change the day-to-day of analysis- Where AI can go wrong in data work, and how to design for trust and review- Why handoff between stakeholders and data teams is still the core bottleneck- What new skills will matter most for analysts and data teams as agentic tools become standard
#datapodcast #analytics #businessintelligence #datatools #datascience #aiforanalytics #agents #agenticanalytics #productanalytics #dataleadership #dataneighbor #livedocs

Explore Data Neighbor free workshops on AI evaluation and agentic analytics (plus our live, hands on course): https://dataneighbor.com/Learn more about Count:https://count.co/In this episode of the Data Neighbor Podcast, we sit down with Ollie Hughes, co founder and CEO of Count, to unpack why traditional BI often fails at the real job teams hire data for: making better decisions, faster.If you care about the future of analytics workflows, this is a must watch.Connect with the team (tell us YouTube sent you!):Shane Butler: https://linkedin.openinapp.co/b02feSravya Madipalli: https://linkedin.openinapp.co/9be8cHai Guan: https://linkedin.openinapp.co/4qi1rConnect with Ollie:https://www.linkedin.com/in/hughesoliver/In this episode, you’ll learn:- Why dashboards are not designed for collaborative decision making- What the “service trap” is and how it keeps data teams in reactive work- How operational clarity helps companies simplify what matters- Why a canvas based interface changes how teams reason with data- What collaborative analysis looks like in practice between analysts and stakeholders- How agile decision making can reduce time to insight and time to decision- How Count thinks about AI agents for analytics without black boxes- What skills will matter most for data teams as AI takes on more of the mechanical work
#datapodcast #analytics #businessintelligence #datatools #datascience #aiforanalytics #agents #agenticanalytics #aievals #productanalytics #dataleadership #dataneighbor #count

Join Our Upcoming AI Evals Cohort! https://maven.com/dataneighbor/ai-evalsSubscribe to our newsletter: https://dataneighbor.substack.com/DNP Swag Store: https://shop.dataneighbor.com/In this episode of the Data Neighbor Podcast, we do something different. No guest. Just the three of us reflecting on what we learned after a year of running the show.We talk honestly about what surprised us most, what turned out to be harder than expected, and why the “real work” of podcasting is everything around the recording: editing, publishing, distribution, and yes, thumbnails.We also dig into what hosting a podcast has done for our careers: better communication reps, more surface area in the industry, and a faster growing network than we expected.Finally, we share a preview of what we are building in 2026, including two big themes we keep seeing across the industry: AI product evaluation and AI powered analytics.If you are thinking about starting a podcast, already building one, or just want a candid behind the scenes look at what it takes to stay consistent for a full year, this episode is for you.Connect with the team (tell us which platform sent you!):Shane Butler: https://linkedin.openinapp.co/b02feSravya Madipalli: https://linkedin.openinapp.co/9be8cHai Guan: https://linkedin.openinapp.co/4qi1rIn this episode, you’ll learn:- What nobody tells you about the time investment behind a podcast- Why consistency is the hardest part, and how to design for it- How having multiple co hosts reduces burnout and improves the show- How podcasting can expand your network and professional opportunities- What we learned from a year of interviewing builders and operators- What we are focusing on in 2026: AI evaluation and AI powered analytics

#podcasting #podcast #creator #contentcreation #careergrowth #networking #aiproduct #aievaluation #analytics #datascience #dataneighbor

In this episode of the Data Neighbor Podcast, we sit down with Shelby Heinecke, PhD, Senior AI Research Manager at Salesforce, to break down what modern AI teams actually look like and how enterprise AI gets built in practice.Shelby shares how her team moves research into production, why small crisp problem definitions outperform ambitious abstractions, and how evaluation before development has become a non negotiable part of the workflow.We also talk about the shifting shape of AI teams, the rising importance of domain experts, and why interdisciplinary collaboration is quickly becoming the core of the field.If you want an inside look at how leading AI orgs actually operate, this is the episode.Connect with the team (tell us which platform sent you!):
- Sravya Madipalli: https://linkedin.openinapp.co/9be8c- Shane Butler: https://linkedin.openinapp.co/b02fe
- Hai Guan: https://linkedin.openinapp.co/4qi1r
Connect with Shelby:https://www.linkedin.com/in/shelby-heinecke/In this episode, you’ll learn:- How enterprise AI research teams actually set direction- Why crisp scope and early evaluation decide which projects reach production- What makes interdisciplinary collaboration essential for AI success- How small models and agents are being deployed across Salesforce- What skills matter most for the next generation of AI roles- Why embodied agents may represent the next major leap in AI
#aipodcast #airesearch #salesforce #aiteams #aiproducts #llm #datascience #mlengineering #aidevelopment #agents #embodiedai #dataneighbor #aifuture