DiscoverAI Loves Data Podcast
AI Loves Data Podcast
Claim Ownership

AI Loves Data Podcast

Author: Data Science Salon

Subscribed: 105Played: 456
Share

Description

The official podcast of Data Science Salon which is now AI Loves Data. We interview top and rising luminaries in data science, machine learning, and AI on the trends and business use cases that are propelling the field forward. The AI Loves Data series is a unique vertical focused conference which brings together specialists face-to-face to educate each other, illuminate best practices, and innovate new solutions in a casual atmosphere with food, great coffee, and entertainment.
70 Episodes
Reverse
Sai works across enterprise AI platforms, agentic systems, RAG, and document intelligence, designing intelligent systems that combine data, reasoning, and automation while remaining scalable, secure, and maintainable.We go beyond the model itself to explore the engineering challenges surrounding modern AI systems, from retrieval and context management to agent orchestration, evaluation, observability, cost, latency, and production reliability.Key Highlights:Why building an impressive AI demo is very different from engineering a system that can reliably support real enterprise users and workflows.The architectural decisions behind retrieval, chunking, reranking, context construction, and reliable enterprise knowledge systems.When agents make sense, how to orchestrate them, and how to avoid unnecessary complexity as agents gain access to more tools and workflows.How teams can measure multi-step AI systems and diagnose whether failures come from retrieval, reasoning, context, tools, or the underlying model.Balancing model capability with latency, infrastructure requirements, and the economics of running enterprise AI in production.How organizations can create maintainable AI platforms without becoming overly dependent on a single model or rapidly changing technology stack.🎧 Tune in to Episode 70 to hear Sai Krishna Reddy Mudhiganti’s perspective on the engineering decisions that separate promising AI prototypes from reliable, production-ready enterprise AI systems.Be sure to mark your calendars for the annual ALD SF on Nov 5. Join us for the Enterprise Builder’s conference to hear from experts on how AI is shaping the enterprise AI. Apply to attend for FREE: https://ailovesdata.com/san-francisco/   Writer and Host: Anna AnisinProduced, edited, and mixed by the Formulatedby Team
AI has changed dramatically over the past seven years. When the Applied AI Summit first launched, practitioners were focused on machine learning, NLP, data science, and the challenge of getting models into production. Today, the conversation has expanded to foundation models, generative AI, RAG, autonomous agents, healthcare AI, and increasingly important questions around evaluation, safety, reliability, and measurable impact. Anna and David explore which technologies genuinely changed the field, what organizations have learned about moving from prototypes to production, and why the industry's focus is increasingly shifting from asking “What can AI do?” to “Can you prove that it works?”Key Highlights:How the industry has evolved from traditional machine learning and NLP to foundation models, GenAI, RAG, and agentic systems.Why creating an impressive AI prototype has become easier while building dependable production systems remains difficult.What organizations have learned about data quality, infrastructure, evaluation, monitoring, workflow integration, and human oversight.Why proving that an AI system actually works has become a defining challenge for practitioners in 2026.Why traditional accuracy metrics aren't enough for generative and agentic systems, particularly in high-stakes industries.What the next three years could bring and which fundamental AI challenges may prove much harder to solve.🎧 Tune in to Episode 69 for a look back at seven years of applied AI and a conversation about what needs to happen next as the industry moves from experimentation toward proving that AI systems are reliable, measurable, and ready for the real world.Register for FREE: https://appliedaisummit.org/  Writer and Host: Anna AnisinProduced, edited, and mixed by the Formulatedby Team
Ritika brings eight years of cybersecurity experience together with hands-on work in LLM security, GenAI applications, AI red teaming, RAG security, and agentic AI risk. Today, her work at the Institute for Security and Technology focuses on emerging AI security risks and translating technical findings into practical approaches to risk management and governance.We explore what changes when AI isn't simply another technology organizations need to protect but becomes a new attack surface capable of accessing data, using tools, interacting with systems, and taking action.Key Highlights:How prompt injection, jailbreaks, adversarial attacks, and other emerging vulnerabilities challenge traditional cybersecurity approaches.What changes when AI systems gain access to tools, data, applications, and the ability to act autonomously.Where vulnerabilities can emerge across retrieval, proprietary data, model interactions, and enterprise GenAI pipelines.Why testing AI systems requires new offensive security techniques and what organizations should actually be testing.How AI can also help defenders automate vulnerability management, compliance, detection, and remediation.Understanding how increasingly autonomous AI systems introduce new questions around monitoring, containment, governance, and human oversight.🎧 Tune in to Episode 68 to hear Ritika Verma's perspective on securing the next generation of AI  and what enterprises need to understand as LLMs and agents become an increasingly important part of the cybersecurity landscape.Be sure to mark your calendars for the annual ALD SF on Nov 5. Join us for the Enterprise Builder’s conference to hear from experts on how AI is shaping the enterprise AI. Apply to attend for FREE:https://ailovesdata.com/san-francisco/   Writer and Host: Anna AnisinProduced, edited, and mixed by the Formulatedby Team
We explore what it takes to turn powerful AI capabilities into products customers actually use and trust, from conversational intelligence and context management to human-AI collaboration, product metrics, and moving beyond AI pilot purgatory.Key Highlights:What separates an AI feature or assistant from a truly agentic product capable of taking action.How product teams should think about context, permissions, user control, reliability, and human oversight.Turning calls, messages, and customer interactions into actionable insights and intelligent workflows.Why promising AI pilots often struggle in production  and what product teams can do differently.Why task completion isn't enough and how teams should think about quality, customer experience, trust, and business outcomes.🎧 Tune in to Episode 67 to hear Tanvi Kopardekar’s perspective on designing the next generation of AI-first products  and what changes when software stops simply answering questions and starts taking action.Be sure to mark your calendars for the annual ALD SF on Nov 5. Join us for the Enterprise Builder’s conference to hear from experts on how AI is shaping the enterprise AI. Apply to attend for FREE:https://ailovesdata.com/san-francisco/   Writer and Host: Anna AnisinProduced, edited, and mixed by the Formulatedby Team
With more than 25 years of experience across real-time machine learning, distributed computing, high-availability systems, and enterprise AI, Ravi brings a deeply technical perspective to one of the biggest questions facing organizations today: How do we move AI agents from impressive demos to systems businesses can actually trust to do work?We explore how enterprises can operationalize AI agents, why industrial systems of record are critical to agentic applications, and what organizations need to consider around context, security, evaluation, governance, and human oversight.Key Highlights:Agentic AI: Hype vs. Reality: Where enterprise AI agents are delivering value today — and where vendor promises are still ahead of what's realistically achievable.From Copilots to AI Workers: What separates a conversational AI assistant from an agent capable of reasoning, taking action, and participating in real enterprise workflows.Agents + Systems of Record: Why integrating AI agents with ERP and other operational systems is essential for reliable, context-aware decision-making.Engineering for Reliability: Lessons from distributed computing, real-time analytics, and high-availability systems that today's agentic AI teams can't afford to ignore.Security, Evaluation & Guardrails: How enterprises should approach permissions, observability, failure detection, governance, and human oversight as AI systems become increasingly autonomous.🎧 Tune in to Episode 66 to hear Ravi Bulusu's perspective on moving beyond the agentic AI hype and building AI systems that are reliable, secure, and ready for the complexity of the real enterprise. Check out: https://theloops.io/ Be sure to mark your calendars for the annual ALD SF on Nov 5. Join us for the Enterprise Builder’s conference to hear from experts on how AI is shaping the enterprise AI. Apply to attend for FREE: https://ailovesdata.com/san-francisco/  Writer and Host: Anna AnisinProduced, edited, and mixed by the Formulatedby Team
loading
Comments