Uploaded June 2026 | Updated November 2026, 33 minutes ago
[2026 - DAY 2 - AI ENGINEERING] I lead large-scale AI implementations and design agents for some of the most security-sensitive enterprises in the world. The pattern is always the same: prototypes and MVPs are easy, running agents in production is not. The moment an agent touches real data and real systems, security, policy, identity, orchestration, and auditability dominate the design. You also discover how many teams must align to make it safe. Without a shared and evolving playbook, the deployment stalls. In this talk I outline the approach I use to build enterprise agents that actually run in production. I cover the core architecture, the supervisory runtime, and the secure execution model that keeps agents predictable: identity boundaries, retrieval controls, RBAC and ABAC alignment, sandboxed tools, and evaluation loops. Through this talk I'd like to show what it really takes to operate agents inside large organizations and provide a practical blueprint you can adapt to your own environment.
The views, opinions, and content shared in this presentation are my own and do not represent those of any employer, past or present. I’d also like to note that my book, Data Engineering for Multimodal AI (O’Reilly), is authored independently and is neutral in perspective. It is not sponsored by Salesforce and does not represent Salesforce’s viewpoints.
SPEAKER:
Vasundra Srinivasan - Director, AI Architect Lead, Salesforce
👉 Sign up for our "No BS" Newsletter to get the latest technical data & AI content: aicouncil.com/newsletter
ABOUT AI COUNCIL:
AI Council brings together the brightest minds in data to share industry knowledge, technical architectures and best practices in building cutting edge data & AI systems and tools.
FIND US:
Website: aicouncil.com
LinkedIn: linkedin.com/company/aicouncilconf
X: https://x.com/aicouncilconf
[2026 - DAY 2 - AI ENGINEERING] I lead large-scale AI implementations and design agents for some of the most security-sensitive enterprises in the world. The pattern is always the same: prototypes and MVPs are easy, running agents in production is not. The moment an agent touches real data and real systems, security, policy, identity, orchestration, and auditability dominate the design. You also discover how many teams must align to make it safe. Without a shared and evolving playbook, the deployment stalls. In this talk I outline the approach I use to build enterprise agents that actually run in production. I cover the core architecture, the supervisory runtime, and the secure execution model that keeps agents predictable: identity boundaries, retrieval controls, RBAC and ABAC alignment, sandboxed tools, and evaluation loops. Through this talk I'd like to show what it really takes to operate agents inside large organizations and provide a practical blueprint you can adapt to your own environment.
The views, opinions, and content shared in this presentation are my own and do not represent those of any employer, past or present. I’d also like to note that my book, Data Engineering for Multimodal AI (O’Reilly), is authored independently and is neutral in perspective. It is not sponsored by Salesforce and does not represent Salesforce’s viewpoints.
SPEAKER:
Vasundra Srinivasan - Director, AI Architect Lead, Salesforce
👉 Sign up for our "No BS" Newsletter to get the latest technical data & AI content: aicouncil.com/newsletter
ABOUT AI COUNCIL:
AI Council brings together the brightest minds in data to share industry knowledge, technical architectures and best practices in building cutting edge data & AI systems and tools.
FIND US:
Website: aicouncil.com
LinkedIn: linkedin.com/company/aicouncilconf
X: https://x.com/aicouncilconf
![LLMs for Data Science
[2025 - Day 1 - Data Science & Algos] Hadley Wickham shares insights from practical applications of LLMs in data science, exploring three key areas where these tools prove genuinely useful beyond the hype: writing code, writing prose, and rectangling non-rectangular data. For data scientists working with text, images, videos, or audio data, this talk offers valuable perspectives on leveraging LLMs effectively for real workflows and transforming fundamentally unstructured information.
ABOUT THE SPEAKER:
Hadley Wickham, Chief Scientist, Posit -
🎟️ GET YOUR TICKET TO AI COUNCIL 2026 🎟️
Meet the worlds top AI infrastructure minds where architects of AI share what works. Three days of high-quality technical talks and meaningful interactions.
→ https://aicouncil.com/sf-2026
⚡ FIND US:
X: https://x.com/AICouncilConf
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
Website: https://aicouncil.com/ LLMs for Data Science](https://i.ytimg.com/vi/MaMQNbwd1MQ/mqdefault.jpg)
![Open Source Success: Learnings from 1 Billion Downloads
[2025 - Day 3 - Lightning Talks] Avi Press shares insights from analyzing over 1 billion open source package downloads across 2000+ projects on Scarf, exploring user behaviors and interactions that reveal best practices for successful open source projects. This talk offers valuable perspectives on download formats, packaging systems, regional trends, and community engagement strategies, essential for maintainers, founders, and executives looking to boost project adoption and drive business growth.
ABOUT THE SPEAKER:
Avi Press, CEO, Scarf -
🎟️ GET YOUR TICKET TO AI COUNCIL 2026 🎟️
Meet the worlds top AI infrastructure minds where architects of AI share what works. Three days of high-quality technical talks and meaningful interactions.
→ https://aicouncil.com/sf-2026
⚡ FIND US:
X: https://x.com/AICouncilConf
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
Website: https://aicouncil.com/ Open Source Success: Learnings from 1 Billion Downloads](https://i.ytimg.com/vi/NFS-xDVVdXE/mqdefault.jpg)
![Real-Time Data Infrastructure and AI: Powering the Next Generation of Analytics
[2025 - Day 2 - Keynote] Aaron Katz and Martin Casado share insights from their pioneering work in database technology and networking innovation, exploring how real-time data infrastructure is fundamentally transforming analytics and AI capabilities. This authentic conversation offers practical wisdom from leaders whove built the technology driving tomorrows data-driven applications, particularly valuable for those scaling high-performance systems or architecting intelligent platforms.
ABOUT THE SPEAKERS:
Aaron Katz, Co-Founder & CEO, ClickHouse
Martin Casado, General Partner, a16z -
🎟️ GET YOUR TICKET TO AI COUNCIL 2026 🎟️
Meet the worlds top AI infrastructure minds where architects of AI share what works. Three days of high-quality technical talks and meaningful interactions.
→ https://aicouncil.com/sf-2026
⚡ FIND US:
X: https://x.com/AICouncilConf
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
Website: https://aicouncil.com/ Real-Time Data Infrastructure and AI: Powering the Next Generation of Analytics](https://i.ytimg.com/vi/NNkrcLQIJrs/mqdefault.jpg)
![The Best Engineer in the Room Doesnt Write Code | Kilo Code
[2026 - DAY 2 - CODING AGENTS] They build systems of agents. In this talk, we explore the shift from session-based AI tools to always-on agents that act, follow up, and own outcomes without waiting to be asked. Well cover what OpenClaw got right, where it falls short, and how were closing the gap with KiloClaw—and what all of this means for the future of how software gets built.
SPEAKER:
Emilie Schario - COO & VP of Engineering, Kilo Code
👉 Sign up for our No BS Newsletter to get the latest technical data & AI content: https://aicouncil.com/newsletter
ABOUT AI COUNCIL:
AI Council brings together the brightest minds in data to share industry knowledge, technical architectures and best practices in building cutting edge data & AI systems and tools.
FIND US:
Website: https://aicouncil.com/
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
X: https://x.com/aicouncilconf The Best Engineer in the Room Doesnt Write Code | Kilo Code](https://i.ytimg.com/vi/O0ELM9ZqkE8/mqdefault.jpg)
![Context Engineering for Video Intelligence: Beyond Model Scale to Real-World Impact | TwelveLabs
[2026 - DAY 3 - LIGHTNING TALK] Video streams combine vision, audio, time-series and semantics at a scale and complexity unlike text alone. At TwelveLabs, we’ve found that tackling this challenge doesn’t start with ever-bigger models — it starts with engineering the right context. In this session, we’ll walk engineers and infrastructure leads through how to build production-grade video AI by systematically designing what information the model receives, how its selected, compressed, and isolated. You’ll learn our four pillars of video context engineering (Write, Select, Compress, Isolate), see how our foundation models (Marengo & Pegasus) and agent product (Jockey) use them, and review real-world outcomes in media, public-safety and advertising pipelines. We’ll also dive into how you measure context effectiveness — tokens per minute, retrieval hit rates, versioned context pipelines — and how this insight drives cost, latency and trust improvements. If you’re deploying AI video solutions in the wild, you’ll leave with a blueprint for turning raw video into deployable insight — not by model size alone, but by targeted context engineering.
SPEAKER:
James Le - Head of Developer Experience, TwelveLabs
👉 Sign up for our No BS Newsletter to get the latest technical data & AI content: https://aicouncil.com/newsletter
ABOUT AI COUNCIL:
AI Council brings together the brightest minds in data to share industry knowledge, technical architectures and best practices in building cutting edge data & AI systems and tools.
FIND US:
Website: https://aicouncil.com/
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
X: https://x.com/aicouncilconf Context Engineering for Video Intelligence: Beyond Model Scale to Real-World Impact | TwelveLabs](https://i.ytimg.com/vi/ODZPMB97DD8/mqdefault.jpg)

![Prompt to Production: Building from Traditional SaaS to True AI-Native SaaS | Webflow
[2026 - DAY 3 - APPLIED AI] Shifting from traditional SaaS to AI-native products demands a new approach to workflows, permissions, and end-to-end product design—while raising the bar for trust, reliability, and organizational readiness. Utkarsh distills lessons from leading AI-native transformations, including how to architect trustworthy AI systems, reorganize teams for high-velocity innovation, and build the cultural momentum required to make AI real inside an organization. Attendees will:
- Learn how to evolve core SaaS architectures into AI-native products without compromising governance or trust
- Gain strategies for reshaping teams, incentives, and hiring practices to accelerate AI adoption
- Explore practical frameworks for driving AI-first transformation across an organization
SPEAKER:
Utkarsh Sengar - VP of Engineering, Webflow
👉 Sign up for our No BS Newsletter to get the latest technical data & AI content: https://aicouncil.com/newsletter
ABOUT AI COUNCIL:
AI Council brings together the brightest minds in data to share industry knowledge, technical architectures and best practices in building cutting edge data & AI systems and tools.
FIND US:
Website: https://aicouncil.com/
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
X: https://x.com/aicouncilconf Prompt to Production: Building from Traditional SaaS to True AI-Native SaaS | Webflow](https://i.ytimg.com/vi/OWL-PZvQjPI/mqdefault.jpg)
![Super-Secret Next Big Thing for DuckDB
[2026 - DAY 1 - DATA ENG & DATABASES] DuckDB has been quietly working on something big. At AI Council 2026, co-creator Hannes Mühleisen finally reveals it.
For years, DuckDB has taken a deliberately embedded, in-process approach to analytical data — powerful, but with one persistent limitation: multiple processes cant write to the same database at once. Until now.
Hannes takes the stage to unveil Quack — a new remote protocol that gives DuckDB client-server capabilities for the first time. With a single extension, any two DuckDB instances on a network can become client and server over HTTP, with the full DuckDB feature set over the wire.
SPEAKER:
Hannes Mühleisen - Co-founder & CEO, Duck Labs
👉 Sign up for our No BS Newsletter to get the latest technical data & AI content: https://aicouncil.com/newsletter
ABOUT AI COUNCIL:
AI Council brings together the brightest minds in data to share industry knowledge, technical architectures and best practices in building cutting edge data & AI systems and tools.
FIND US:
Website: https://aicouncil.com/
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
X: https://x.com/aicouncilconf Super-Secret Next Big Thing for DuckDB](https://i.ytimg.com/vi/O_tzHeDpjrE/mqdefault.jpg)


![A Modern Data Stack in Healthcare
[2025 - Day 3 - Lightning Talks] Anil Sadineni shares insights from applying modern data stack technologies to healthcare data exchange, exploring how FHIR convergence with advances in data processing creates opportunities to transform healthcare data flows. This talk offers valuable perspectives on revising socio-technical assumptions for decentralized health data exchange, essential for data engineers and healthcare data enthusiasts working with complex regulatory requirements across payer and provider entities.
ABOUT THE SPEAKER:
Anil Sadineni, Principal Software Engineer, 1uphealth -
🎟️ GET YOUR TICKET TO AI COUNCIL 2026 🎟️
Meet the worlds top AI infrastructure minds where architects of AI share what works. Three days of high-quality technical talks and meaningful interactions.
→ https://aicouncil.com/sf-2026
⚡ FIND US:
X: https://x.com/AICouncilConf
LinkedIn: https://www.linkedin.com/company/aicouncilconf/
Website: https://aicouncil.com/ A Modern Data Stack in Healthcare](https://i.ytimg.com/vi/PfZZ1qkfTsU/mqdefault.jpg)