←
Data Science Dojo60,165 posts
Data Science Dojo avatar

Data Science Dojo

@DataScienceDojo

We make learning data science and LLMs easy! Join the community of 10,000+ professionals. #DSDojo

◷ Joined March 2013▧ 12,982 Photos and videos
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Jun 30

The July 14th cohort is coming up soon. 10 weeks. 12,000+ alumni. Spots are filling fast. The Agentic AI Bootcamp is back and it covers the full stack of what matters right now. Week 1, you're wrapping your head around how Transformers actually work. Week 3, you're building with LangChain and setting up your first vector database for a RAG pipeline. By week 5, you're not just prompting. You're engineering context and designing systems using Agentic design patterns used in production. Week 7 hits and you're implementing Model Context Protocol and learning how agents actually talk to each other. Week 10? You ship a multi-agent LLM application. From scratch. In 10 weeks. That's what the Agentic AI Bootcamp cohort starting July 14th looks like. 3 hours a week, structured to take you from foundations to deployment. Register now before it's too late: hubs.la/Q04n3Xcm0 #agenticai #rag #aibootcamp #vectordatabases

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 10

🔴 LIVE NOW: Allison Ding, Senior Developer Advocate at NVIDIA, is leading the hands-on tutorial “Teaching Coding Agents to Think in GPU: Writing, Testing, and Shipping Agentic Skills” at the Agentic AI Conference. 🖥️ Ask a coding agent to speed up a slow pandas job and you often get pandas back: correct, confident, and still on the CPU. In this 45-minute session, you’ll watch that failure happen live, take apart a real production skill for NVIDIA CUDA-X Data Science, then write one together, and learn how to test it against the same task without it. Agentic skills are plain markdown files that route an agent to the right library, ground it in the current docs, and validate its work at runtime. You’ll leave with a working skill and a repeatable way to prove it works. Allison helps developers get more out of GPU-accelerated AI libraries and tools, with a focus on accelerated data science and large language models, and has more than a decade of industry experience. ▶️ Join live on Zoom: us02web.zoom.us/webinar/regi… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 10

🔴 LIVE NOW: Benjamin Squire, Senior Developer Advocate at Neo4j, is leading the hands-on tutorial “Making a Data Agent Trustworthy with a Semantic Layer and Auditable Memory” at the Agentic AI Conference. 🔍 Data agents rarely fail loudly. They return plausible numbers, query data they shouldn’t, and leave no trail you can audit. In this session, you’ll run one text-to-SQL agent over a realistic 33-table BigQuery warehouse twice: once with the usual list-tables-and-guess tooling, and once with graph-based context from two Neo4j Labs projects. Neocarta adds a semantic layer of governed metric definitions, business synonyms, and sensitivity tags, checked before any query runs. Neo4j Agent Memory records each reasoning step and what it touched, so “who accessed compensation data, and why?” becomes a single graph query. Then comes the hard part: a prompt injection planted in a support ticket. Whether or not the model takes the bait, you’ll trace a poisoned team note to its source, find every session it influenced, and quarantine it. You’ll leave with a public repo to rerun everything yourself. ▶️ Join live on Zoom: us02web.zoom.us/webinar/regi… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 10

🔴 LIVE NOW: “The attack is already in your index” is happening right now at the Agentic AI Conference, with Scott Askinosie (Principal AI Architect, SteelEngine), James Luan (CTO, Zilliz) and Kiran Bondalapati (Engineering Leader, Glean). 🎙️ In 2025, researchers embedded instructions in an ordinary web form and watched an agent retrieve sensitive CRM records and send them to an external address. Nothing about the model was compromised: the payload arrived the way all context does. As agents pull from documents, tool responses, and memory, that content becomes executable in a way most security reviews still don’t treat it. Our panelists are covering what teams are actually putting in place: provenance tracking, validation of external inputs, and detection for content nobody wrote by hand. Can poisoned context be reliably caught and cleaned, or is the only durable answer keeping untrusted content away from anything privileged? Scott is Principal AI Architect for Knowledge and Memory Systems at SteelEngine. James is the CTO of Zilliz, Chair of the Milvus community, and a member of the LF AI & Data Foundation’s Technical Advisory Committee. Kiran builds enterprise AI systems with trusted context, permissions, and agentic workflows at Glean, and was previously VP of Engineering at TRM Labs. ▶️ Join live on Zoom: us02web.zoom.us/webinar/regi… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 10

🔴 LIVE NOW: Iswarya Alex (Senior Staff Software Development Engineer, AMD) and Satya Devineni (Product Application Engineer, AMD) are leading the hands-on tutorial “Build Secure Always-On Local Agents on AMD Ryzen™ AI PC using Lemonade” at the Agentic AI Conference. 🛠️ An always-on agent is only worth having if you can trust it with real work. In this hands-on workshop, you’ll build one on a Ryzen™ AI PC that runs day and night, works on your inbox and calendar, and stays inside limits you set yourself. Lemonade Server keeps every model call local across the NPU and GPU, so nothing leaves the machine. You’ll put the agent in a sandbox where what it can read, run and reach is a policy you write and can change, and use it to build a fun productivity use case. ▶️ Join live on Zoom: us02web.zoom.us/webinar/regi… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 10

🔴 LIVE NOW: “Every sandbox ends up with exceptions” is happening right now at the Agentic AI Conference, with Suman Debnath (Director of Developer Relations Engineering, Crusoe), Dwayne McDaniel (Principal Developer Advocate, GitGuardian) and Ori Bendet (VP of Product, Docker). 🎙️ The more an AI agent can do, the more useful it becomes, and the more damage it can cause through a bug, a bad instruction, or a prompt it should never have trusted. Telling an agent not to do something is not the same as preventing it, so every team deploying agents against tools, code, or infrastructure eventually faces the same question: where does the boundary go? Our panelists are digging into the isolation architectures teams have actually shipped, from container-level isolation and scoped credentials to network egress rules and approval gates on destructive operations, along with the failures that shaped them. Does a well-designed sandbox genuinely contain an agent, or does every production deployment collect exceptions until the boundary is mostly decorative? Suman focuses on AI infrastructure and managed inference at Crusoe. Dwayne specializes in secrets security across the full credential layer. Ori leads Docker Hub and developer tooling for more than 20 million developers. ▶️ Join live on Zoom: us02web.zoom.us/webinar/regi… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 10

🔴 LIVE NOW: “How do agents work, and how should they work?” is happening right now at the Agentic AI Conference, with the RAG Pack: Luis Serrano (Founder, Serrano Academy), Chris McCormick (AI Researcher and Educator, InnerWorkings AI), Jay Alammar (Director and Engineering Fellow, Cohere) and Josh Starmer (Founder, StatQuest), moderated by Raja Iqbal (Founder, Ejento AI). 🎙️ Everyone is building agents, but few people can say what actually happens between a prompt and a finished task. Under the hood, an agent is a language model running in a loop: it reads its context, decides on a step, calls a tool, reads the result and decides again. Retrieval, memory and planning all shape what goes into that loop, and most agent failures trace back to one of them. The RAG Pack is taking the agent apart piece by piece: how the model chooses its next action, what retrieval and memory really add, how reinforcement learning is changing the way models learn to use tools, and why agents that impress in a demo often stall in practice. Then comes the harder question: when is an agent the right design, when does a simple pipeline win, and how do you know yours is getting the job done? ▶️ Join live on Zoom: us02web.zoom.us/webinar/regi… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 9

🔴 LIVE NOW: Paige Bailey, AI Developer Relations Lead at Google DeepMind, is presenting right now at the Agentic AI Conference. 🚀 What’s new and what’s next from Google DeepMind? Join Paige for a first-hand look at the latest models and agentic workflows coming from Google DeepMind. Here’s what you’ll take away. ✅ Live demos with Gemini 4, NanoBanana 2.1, Gemini TTS, and more ✅ A first-hand look at the agentic workflows coming from Google DeepMind ✅ Hands-on coding patterns for going from prototype to production with next-gen autonomous systems Paige has helped define the developer experience for Gemini, PaLM 2, and Gemma, and presented AI Studio, the Gemini APIs, and new additions to Antigravity in the developer keynotes at Google I/O ’25 and ’26. ▶️ Watch live on YouTube: youtube.com/watch?v=72NuTUHA… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachmentPost attachmentPost attachmentPost attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 9

🔴 LIVE NOW: “Teammates, not tools” is happening right now at the Agentic AI Conference, with Raja Iqbal (Founder, Ejento AI), Sam Partee (Co-founder and CTO, Arcade), Adnan Ijaz (Senior Director of Product Management, Agentic Data Cloud, Google) and Jamil Valliani (Head of Product, AI, Atlassian). 🎙️ Companies already know how to bring a person onto a team: a defined role, access scoped to that role and a clear line on what needs a manager’s sign-off. Yet many hand AI agents the power to move money and rewrite customer records through shared service accounts, with permissions nobody scoped and no owner named anywhere. This panel treats agents as teammates who need onboarding, not software that needs installing. Our panelists are getting concrete on per-agent identity, approval thresholds and evidence trails that hold up when a regulator asks who authorized what, and on how much authority a digital teammate should get on day one. Raja co-authored the Harvard Business Review article this panel is built on. Sam builds the runtime that lets AI agents take real actions across enterprise systems securely. Adnan leads product for Google’s Agentic Data Cloud. Jamil leads product for AI at Atlassian, after more than 15 years leading product teams at Microsoft. ▶️ Watch live on YouTube: youtube.com/watch?v=72NuTUHA… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 9

🔴 LIVE NOW: “Secure AI needs forward deployed engineers” is happening right now at the Agentic AI Conference, with Aater Suleman (Co-founder, Vixul), Ben Wilson (Databricks) and Mark Ghannam (LandingAI). 🎙️ A vendor’s security guidance assumes a reference architecture nobody actually runs. A compliance framework assumes clean data ownership and a documented approval chain. Each arrives as a document, and every organization receiving it has its own legacy systems, its own risk appetite and its own way of deciding things. The distance between the guidance and the deployment is where AI programs stall. Forward deployed engineers close that distance. They embed with a customer and turn guardrails, threat models and control requirements into integrations that fit how that enterprise actually runs. This panel is about what the work really takes and whether it scales, with real disagreement on the question underneath it: is the FDE a permanent fixture of enterprise AI, or a sign the tooling isn’t finished? ▶️ Watch live on YouTube: youtube.com/watch?v=72NuTUHA… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 9

🔴 LIVE NOW: Jaya Vivekanandan, Applied AI Engineer at LandingAI, is leading the hands-on tutorial “Show Me Where It Says That: Agents That Cite Every Word in a Document” at the Agentic AI Conference. 🛠️ Every company has documents. Can an AI platform be trusted to read them for you? In this session you’ll work with DPT-3 Verity in LandingAI’s Agentic Document Extraction, a parsing model that grounds every word it reads, in text and in tables, to its exact spot on the page and gives it a confidence score. Then you’ll build a document version-control pipeline that parses several revisions of a lease, lines up the words and finds every addition, removal and edit, with each change highlighted on the original pages. You’ll also learn why a low confidence score means “check this,” not “this is wrong.” Some Python helps, and no document AI experience is needed. ▶️ Watch live on YouTube: youtube.com/watch?v=72NuTUHA… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment
Data Science Dojo avatar
Data Science Dojo@DataScienceDojo·Nov 9

🔴 LIVE NOW: “How sovereign is sovereign AI?” is happening right now at the Agentic AI Conference, with Hamza Farooq (Founder, Traversaal), Malte Pietsch (Co-founder and CTO, deepset) and Zain Hasan (Staff AI/ML Engineer, Together AI). 🎙️ A European bank runs a sovereign cloud region, hosts its model weights domestically and keeps every byte of customer data inside the bloc. Yet the chips came from one vendor, the architecture from a foreign lab’s published research, and the inference software from a company subject to its own government’s powers. Something real has been achieved, and something important has not. Our panelists are separating the three layers that usually get collapsed into one, data, models and compute, and asking what hosting your own weights, data and compute actually buys you. Hamza leads forward-deployed engineering for Fortune 100 companies after close to 20 years building large-scale systems at Google and Walmart. Malte’s company, deepset, is behind the open-source Haystack framework and focuses on AI for regulated industries and the public sector. Zain is a widely followed AI educator whose DeepLearning.AI courses have reached more than 200,000 learners. ▶️ Watch live on YouTube: youtube.com/watch?v=72NuTUHA… #datasciencedojo #agenticaiconference #futureofdataandai

Post attachment