Analytics Vidhya

We're excited to welcome Bharath Kumar Bolla, Associate Director of MLOps and Data Science at Novartis, to DataHack Summit 2026. As AI agents become more autonomous, one question matters more than ever: Why did the model make that decision? In his hack session, Opening the Black Box of AI Agents: Mechanistic Interpretability, Bharath takes us inside the inner workings of transformers to understand how models store knowledge, reason, and sometimes fail: analyticsvidhya.com/datahacksummit… Join Bharath at DHS 2026 and learn how to make AI agents more transparent, debuggable, and trustworthy.

We're excited to welcome Arun Prakash Asokan, Global Director of Data Science and AI at Novartis, to DataHack Summit 2026. With nearly two decades of experience in AI and a track record of driving enterprise-scale AI adoption, Arun has been at the forefront of translating cutting-edge AI research into real business impact. In his workshop, "When Agents Meet Graphs: The Rise of Knowledge-Driven GenAI," he will explore how knowledge graphs and agentic systems can move beyond traditional RAG to enable deeper reasoning, richer context, and more intelligent AI applications. If you're interested in the next evolution of GenAI systems, this workshop is not to be missed. 📅 Aug 05–08 | Bengaluru | analyticsvidhya.com/datahacksummit…

We're delighted to welcome Saurav Agarwal, Senior Manager - Solutions Architecture and Engineering at NVIDIA, to DataHack Summit 2026. With 14+ years of experience in Generative AI, cloud computing, and large-scale AI systems, Saurav helps organizations build secure, scalable, and production-ready AI solutions. In his hack session, "OpenClaw: To Deploy Autonomous AI Assistants Safely and at Scale," he'll explore how sandboxing, policy guardrails, and governance can make enterprise AI agents secure, reliable, and ready for real-world deployment. 📅 Aug 05–08 | Bengaluru | analyticsvidhya.com/datahacksummit…

Want to understand how AI bridges the gap between text and vision? 👁️ 🙌This deep dive breaks down Modern Vision Language Models (VLMs), how they work, and why they are transforming multimodal AI. Read the full breakdown on @AnalyticsVidhya analyticsvidhya.com/blog/2026/07/m…

🚀 xAI has launched Grok 4.5, its most capable model yet for coding, AI agents, and knowledge work. Highlights: ⚡ Up to 80 tokens/sec 🧠 Builds complete apps from a single prompt 📊 Creates PowerPoint, Excel, and Word docs 💰 Uses 4.2× fewer output tokens than Opus 4.8 on SWE-Bench Pro, making long coding workflows more cost-efficient. Read more: x.ai/news/grok-4-5 #AI #GenerativeAI #Grok45 #xAI #Coding

We're delighted to welcome Raji Karat, Head and Assistant Vice President of Research & Development at TNQTech, to DataHack Summit 2026. With over 20 years of experience in scholarly publishing, Rajalakshmi has been driving AI, automation, and content intelligence at scale, combining AI with structured data, domain expertise, and human judgment. In her PowerTalk, "Human × AI: Turning Content into Intelligence," she'll explore how organizations can transform unstructured content into trustworthy, knowledge-rich AI systems that go beyond retrieval to true reasoning 📅 Aug 05–08 | 📍Bengaluru | analyticsvidhya.com/datahacksummit…

We're delighted to welcome Saurav Agarwal, Senior Manager - Solutions Architecture and Engineering at NVIDIA, to DataHack Summit 2026. With 14+ years of experience in Generative AI, cloud computing, and large-scale AI systems, Saurav helps organizations build secure, scalable, and production-ready AI solutions. In his hack session, "OpenClaw: To Deploy Autonomous AI Assistants Safely and at Scale," he'll explore how sandboxing, policy guardrails, and governance can make enterprise AI agents secure, reliable, and ready for real-world deployment. 📅 Aug 05–08 | Bengaluru | analyticsvidhya.com/datahacksummit…

We're excited to welcome Vignesh Kumar, Director of AI Engineering at Ford Motor Company, to DataHack Summit 2026. Vignesh will be joining our Hack Panel: Battle of Coding Agents to discuss one of the hottest topics in AI right now: ⚔️ Claude Code vs Codex vs Antigravity ❗️Which coding agent fits best into real engineering workflows? ❗️Where does each one shine? And how are they changing the way software gets built? With deep expertise in AI engineering and production AI systems, Vignesh brings a practitioner's perspective to the future of AI-native development. Join us at DHS 2026 for the battle of coding agents: analyticsvidhya.com/datahacksummit…

Google DeepMind just published a roadmap on what happens after AGI. Led by Shane Legg (DeepMind co-founder) and Marcus Hutter (creator of the AIXI theory), the report maps out how AI could progress from human level intelligence (AGI) to artificial superintelligence (ASI), intelligence that surpasses even large human organisations, not just individuals. Four pathways they outline: ➡️ Scaling current AGI systems further ➡️ A fundamental paradigm shift in AI architecture ➡️ Recursive self improvement, AI improving itself ➡️ ASI emerging from large networks of AI systems working together The report does not predict a date. Instead it builds a framework using the Legg-Hutter score and AIXI (a theoretical "maximum intelligence" agent) to reason about how fast or slow this transition could be, and what could bottleneck it. Worth a read if you are thinking about where AI development goes after AGI. Read the full paper here: arxiv.org/abs/2606.12683

Want to understand how AI bridges the gap between text and vision? 👁️ ✍️ This deep dive breaks down Modern Vision Language Models (VLMs), how they work, and why they are transforming multimodal AI. Read the full breakdown on @AnalyticsVidhya 👇 analyticsvidhya.com/blog/2026/07/m… #AI #MachineLearning #VLMs #ComputerVision #GenAI

Vector RAG retrieves information by text similarity, while GraphRAG retrieves it by explicit data structure. 🚀 🤖 We built both systems on the exact same dataset to map out where simple semantic search breaks and where 🗺️ entity maps dominate global, corpus-wide queries. See the performance benchmarks and Python tutorial: analyticsvidhya.com/blog/2026/06/g… #GraphRAG #VectorRAG #RAG #GenerativeAI #LLMs #AnalyticsVidhya

The biggest difference between these plans isn’t what they can do, it’s what happens when you push them to the limit. 🛠️📊 When you hit your usage limits during long, intense work sessions, the platforms handle it completely differently. The threshold breakdown: 🛑 Claude Pro: Hits a hard wall with no automatic fallback model tier. 📉 ChatGPT & Gemini: Smoothly drop you down a tier so your chat flow never fully breaks. 💻 Coding Setup: Claude and ChatGPT run tight file loops, while Gemini excels with massive text contexts. Find your perfect match: analyticsvidhya.com/blog/2026/06/c…

Stop baby-sitting your coding LLMs. It's time for models that can fix themselves. 💻🛠️ In our testing, Claude Sonnet 5 didn't just patch errors blindly. It diagnosed root causes, wrote verification tests, and ran code to confirm its own fixes worked. What you need to know about the launch: 🏷️ The Default Model: It is now completely free and live for all standard accounts. 📉 Slash Running Costs: API pricing is down to $2/1M input tokens through August. 🛡️ Fewer Hallucinations: It aggressively screens bad inputs and cuts down on made-up facts. See the full code test and benchmarks: analyticsvidhya.com/blog/2026/07/c…

Excited to have Pranjal Singh, Staff Data Scientist at udaan.com, joining us at DataHack Summit 2026! 🎉 At DHS 2026, Pranjal leads a beginner-friendly hack session, Prompt & Inference Optimization: Building Reliable AI Systems. 🔍 What you'll learn: ✨ Why prompt hacking doesn't scale, and what to do instead ✨ Core concepts: signatures, modules, optimizers, evaluation ✨ Techniques to improve accuracy, consistency, latency, and cost ✨ Tackling hallucinations, brittle prompts, and scaling issues in production ✨ A live demo building and optimizing an AI pipeline end to end 📅 Aug 05–08 | 📍Bengaluru | analyticsvidhya.com/datahacksummit…

We're excited to welcome Arun Prakash Asokan, Global Director of Data Science and AI at Novartis, to DataHack Summit 2026. With nearly two decades of experience in AI and a track record of driving enterprise-scale AI adoption, Arun has been at the forefront of translating cutting-edge AI research into real business impact. In his workshop, "When Agents Meet Graphs: The Rise of Knowledge-Driven GenAI," he will explore how knowledge graphs and agentic systems can move beyond traditional RAG to enable deeper reasoning, richer context, and more intelligent AI applications If you're interested in the next evolution of GenAI systems, this workshop is not to be missed. 📅 Aug 05–08 | Bengaluru | analyticsvidhya.com/datahacksummit…

We're excited to welcome Abhishek Kumar, Senior Director of Data Science at @PublicisSapient, to DataHack Summit 2026. With nearly two decades of experience in AI and Machine Learning, Abhishek has helped enterprises move from AI experimentation to production-scale systems that deliver real business value. In his hack session, "Harness Engineering for Enterprise AI Agents," he'll explore what separates impressive AI demos from reliable, governable, and production-ready enterprise agents. If you're building AI agents for the real world, this session is not to be missed. 📅 Aug 05–08 | 📍Bengaluru | analyticsvidhya.com/datahacksummit…

Using AI daily doesn't mean you have to give up your digital privacy. It just requires basic hygiene. 🧠🛡️ This quick guide cuts through the noise to show you exactly how to secure your accounts without breaking your daily workflow. What you will master in minutes: 🔀 Training Toggles: Turn off training data collection in under 30 seconds. ⏳ Incognito Chats: Run temporary conversations that leave zero footprint. 📦 Local Alternatives: When to swap cloud apps for completely offline, local models. Link: analyticsvidhya.com/blog/2026/06/a… #AIChatbots #Privacy #DataSecurity #ChatGPT #ClaudeAI #AnalyticsVidhya

Excited to welcome Vikas Agrawal, Director of Data Science in AI and Applied Science at Oracle, to DataHack Summit 2026. His work sits at the intersection of AI agents, domain knowledge, LLMs, and real-world enterprise decision-making, tackling one of the most important challenges in AI today: creating intelligent systems that can solve complex problems at scale. Join us at hashtag#DHS2026 to learn from one of the leaders shaping the next generation of enterprise AI. 📅 Aug 05-08 | 📍 Bengaluru | analyticsvidhya.com/datahacksummit…

Stop manually building, debugging, and maintaining custom multi-agent routing pipelines. 🛠️⚡ Sakana Fugu introduces "Multi-Agent System as a Model." You send one request to an OpenAI-compatible API, and Fugu handles the coordination. How the tiers stack up: ⚖️ Fugu (Standard): Balanced speed and performance for everyday coding. 🚀 Fugu Ultra: Maximum quality pooling to tackle high-stakes, multi-step problems. Get the code & benchmarks: analyticsvidhya.com/blog/2026/06/s…

We're excited to welcome Mathangi Sri Ramachandran Sri, Co-founder of YuVerse, to DataHack Summit 2026. From leading AI and data initiatives at Yubi, Citibank, HSBC, GE, PhonePe, and Gojek to building AI-powered solutions at YuVerse, Mathangi has spent her career turning data into business impact. She is the author of two books, holds over 100 patents, and is a strong advocate for applied AI in the enterprise. Join us at DHS 2026 to learn how organizations can use AI to drive smarter decisions, better customer experiences, and measurable outcomes at scale. 📅 Aug 05–08 | 📍Bengaluru | analyticsvidhya.com/datahacksummit…