The Ravit Show

From the floor of Data Streaming Summit 2025, I had the chance to sit down with Ashwin Raja from Motorq, a StreamNative customer, to talk about the real-world impact of streaming.Ashwin shared how his team turns live vehicle signals into immediate decisions, where waiting even a few minutes is not an option. Batch systems couldn’t keep up, so Motorq leaned into StreamNative to build reliable pipelines, manage costs, and move from event to action with confidence.We spoke about what excites him most over the next year: the intersection of streaming and AI, smarter enrichment, anomaly detection, and agents acting safely on live data. The Streaming Lakehouse vision also stood out, making analytics-ready data available without extra hops.Ashwin appreciated the open, multi-technology approach of this summit, which mirrors how real systems get built. His takeaways and reflections are in the episode now, watch the full conversation to hear how Motorq is shaping the future of connected-vehicle data with streaming.#data #ai #datastreamingsummit2025 #streamnative #theravitshow

AI that can actually help you build. I had a blast interviewing Vojta Tuma, Field CTO at Keboola, on The Ravit Show, we discussed AI assistants and MCP for data engineering!Here are a few key takeaways — • MCP turns a chatbot into a teammate. It can use tools, call APIs, and take actions. That is why it matters for engineers. • New workflows open up. You can sketch a pipeline in conversation, run a job, debug a failure, or trigger a transform from one place. • Safety is built in. Step by step permissions. Full logs. Reviews and approvals before changes go live. Propose a fix, dry run it, then merge. • The result is faster work without losing control. Less glue code. More time on the real problems.Let developers, analysts, and data engineers ship faster—AI handles the heavy lifting, Keboola keeps it reliable -- https://bit.ly/46Fkvt1Don’t forget to follow Keboola!!!!#data #ai #keboola #mcp #theravitshow

Last week at BDL, we caught up with my friend Pavel Dolezalžal, co-founder at Keboola - and let me tell you, this conversation was a ride.We started with Keboola Agents, which are already live and helping data teams debug pipelines, document, and automate safely inside a governed platform.Then Pavel dropped the big news: An open-source, conversational ETL pipeline generator - Osiris.You literally describe the problem, and Osiris drafts a transparent, deterministic YAML pipeline.You review, approve, commit - and it just runs. No black boxes, no daily AI gambling.That would be a big shift: AI proposes. Humans approve. Execution stays auditable.Too bold? Have a look for yourself!Links & Resources- Learn more about Keboola Agents: http://bit.ly/4pL6a7h- Explore Osiris on GitHub: https://github.com/keboola/osiris - Connect with Pavel on LinkedIn: https://www.linkedin.com/in/paveld/#data #ai #agents #keboola #theravitshow #dataengineering

I’m at Big Data London and just wrapped a conversation with Jordan Burger, Applied AI Lead at Keboola. He’s focused on building AI that data teams can trust in production, and he’s one of the builders behind Keboola Data Agents.Here’s what we got into:* Why a full data platform matters when AI needs real context, not just a few tables* The lifecycle context agents need that standalone tools never see* What breaks when AI only sees fragments of the data story* How platform-native AI handles messy data, lineage, policies, and tribal knowledge* Why context-aware implementation can matter more than the model itself* The role of historical data stores for agents and what this means for future architecturesJordan is on stage tomorrow to showcase this work. Both agents are live on the Keboola platform. Link in the description to spin up your first flow with your own data.If you’re at Big Data London, meet the Keboola team at Booth L40.#data #ai #bigdataldn2025 #keboola #theravitshow

Speed to insight wins. If analytics does not move faster than the business, it gets ignored.I caught up with the amazing Deborah from Alteryx at Big Data London to talk about what is working right now.We spoke about — • The floor is hands on this year. Fewer slides, more real builds • Alteryx One fits that mood. One place to prep, analyze, automate, and share so teams move faster • Customers are turning hours into minutes on routine work. More people can self serve without waiting in line • The edge is in clean handoffs. Less copy and paste, clearer ownership, smoother reviews • Next up is scale with control. Bring governance into the same flow so speed does not slipWe will keep the conversation going! Don’t forget to follow Deborah for the great work she does in the Data & AI space!!!!#data #ai #alteryx #theravitshow

Orchestration is where AI work meets reality. If it fails, everything slows.I caught up with Luke Lipan, Sales Leader for North America at Kestra, to get the view from the field.Here’s the signal:Execs start with risk and reliability. Is it visible. Is it repeatable.Teams love open source to move fast. Scale, support, and ownership tip them to enterprise.Security and governance are not add-ons. They decide the deal.Success is measured in fewer incidents, faster cycles, and cleaner handoffs. Cost is a result, not the target.Consolidation is happening. One platform beats a patchwork when the stakes rise.If you are at Big Data London, meet the Kestra team at Booth N48.#data #ai #bigdataldn2025 #kestra #theravitshow

Declarative beats duct tape. I spoke to Eddie Jaoude, Lead Education Engineer at Kestra at the Big Data London, to unpack how teams actually ship with modern orchestration.Here is the flow we covered:* Why defining workflows declaratively changes the game for developers coming from Python DAGs* How teams adopt Kestra step by step without a risky forklift migration* Using the Playground in Kestra 1.0 to test steps safely and tighten the feedback loop* Plugging into existing stacks with plugins and a language-agnostic design so you do not rewrite everything* The first usage patterns Eddie sees in the wild, from scheduled jobs and data pipelines to event-driven tasksClear takeaways. Fewer moving parts. Faster iteration. Stronger reliability.#data #ai #kestra #bigdataldn #theravitshow

Everyone is racing to build AI. Almost no one is securing it end to end. I sat down with Vidya Shankaran, CISSP, Field CTO, Commvault and we talked about the real picture. AI risk is not just about models. It is about data, access, and clean recovery. Most teams are missing the biggest gaps.We covered:• The top AI security threats right now and where teams underestimate risk• The AI stack no one is securing in practice• If a CISO asks where to start and how to prioritize controls• Why traditional data access governance is broken• What is at stake if enterprises do not modernize access to sensitive data• How Satori delivers faster access with tighter control• How Commvault protects AI end to end and even recovers vector indexes and configs after an incidentWe also talked about SHIFT 2025 in New York on Nov 11–12. This event will bring together AI security, data access, and resilience with real answers.In-person: https://lnkd.in/dZ6t8nbY?If you cannot attend in person, there is a full virtual experience on Nov 19.Virtual: https://lnkd.in/dz8yhf-cThis was a raw and tactical conversation. If you care about building AI that moves fast with control, you should watch it.SHIFT will set the tone for how enterprises secure AI in 2025.#data #ai #security #shift2025 #theravitshow

A candid conversation with Navin Budhiraja, CTO and Head of Products at Vianai Systems, Inc. on The Ravit Show in Palo Alto. From Bhilai to Palo Alto. Navin topped the IIT entrance exam, studied at Cornell, and led at IBM, AWS, and SAP. We sat down to talk about building AI that enterprises can actually use.What we covered:- Vianai’s mission and the hila platform: why it exists and the problem it solves- How hila turns enterprise data into something teams can interact with in plain language- Responsible AI in practice: tackling hallucinations and earning trust- Why a platform like hila is needed even with powerful foundation models- Conversational Finance: what makes it useful for finance teams- Real integrations: ERP, CRM, HR systems, and how this works end to end- Security for the real world: air-gapped deployments, privacy, and certifications- The road ahead: how AI, IoT, and cloud are converging in the next 2 to 3 years- Advice for the next generation of builders from Bhilai, the IITs, and beyondWhy this matters:Enterprises want outcomes, not hype. Navin’s lens on trust, flexibility, and scale shows how AI moves from pilot to production.Thank you, Navin, for the clear thinking and straight answers. Full interview on The Ravit Show YouTube channel is live.#data #ai #vianai #theravitshow

Stop moving data. Start shipping models. I sat down with Molly Presley from Hammerspace and David A. Chapa from Hitachi Vantara to unpack the new Hitachi iQ M Series and how it pairs with Hammerspace for real GenAI work.What we cover• Running AI on data where it lives across sites and clouds• One global namespace so teams stop copying data back and forth• Flexible GPU choices that fit your budget today and scale later• Independent scaling of compute and storage to control costs• How this shows up on day one for training and inferenceWhy this matters• Faster first results without re-architecting your data• Fewer bottlenecks keeping GPUs idle• A simple path from pilot to productionWatch the full conversation. Link in the first comment.Thank you Molly and David for the clear breakdown and the practicality teams need right now.#data #ai #hammerspace #theravitshow