Agents sound smart until millions of users show up. A real talk on tools, UX, and why autonomy is overrated.

Tool definitions are the new Prompt Engineering

As AI applications move beyond rows and columns into images, video, embeddings, and graphs, traditional query languages like SQL and Cypher start to crack. This post explains why ApertureDB chose to design a JSON-based query language from scratch—one built for multimodal search, data processing, and scale. By aligning with how modern AI systems already communicate (JSON, agents, workflows, and natural language), ApertureDB avoids brittle joins, performance tradeoffs, and DIY pipelines, while still offering SQL and SPARQL wrappers for familiarity. The result is a layered, future-proof way to query, process, and explore multimodal data without forcing old abstractions onto new problems.

Beyond SQL: The Query Language Multimodal AI Really Needs

Everyone’s arguing about agents. Jonathan Wall says the real fight is about sandboxes, isolation, and why most “agent platforms” are doing it wrong.

The Future of AI Agents is Sandboxed

MCP lets your agents connect to Slack, GitHub, your database, and whatever else you throw at it. Great for productivity. Terrible for security. When an agent can call any tool through any protocol, you've got a problem: who's actually making the request? What can they access? And when something breaks—or gets exploited—how do you even trace it back? This talk covers what breaks when agents go multi-protocol: authentication that doesn't account for agent delegation, permission models designed for humans not bots, and audit trails that disappear when Agent A spawns Agent B to call Tool C. I'll walk through real attack scenarios—prompt injection leading to unauthorized API calls, credential leakage across protocol boundaries, and privilege escalation through tool chaining. Then we'll dig into what actually works: identity verification at protocol boundaries, granular permissions that follow context, not just credentials, and audit systems built for non-human actors. You'll leave knowing how to implement MCP without turning your agent system into an attack surface, and what to build (or demand from vendors) to keep agent-to-tool communication secure.

MCP Security: What Happens When Your Agents Talk to Everything? // Rosemary Nwosu Ihueze  

<h1>The 10 Most Important Developments in AI in 2025 - Deep-Dive Roundtable for Practitioners</h1><h2>[Special 2-Hour Session]</h2><h2>To close out the year, we’re hosting a special extended Reading Group session, a two-hour, practitioner-level deep dive into the 10 most consequential developments in AI in 2025.</h2> This isn’t a recap for casual observers.This is a technical, industry-facing breakdown of what actually changed this year, why it matters, and what engineers should expect heading into 2026.We’ve curated the top developments with input from the community, and assembled an expert panel of engineers, researchers, and builders to go deep on their real-world impact.What We’ll Cover, A Sneak PeekExpect sharp commentary, architecture-level insights, and grounded predictions on topics such as:<ol><li data-list="bullet">MCP servers and the rise of coding agents</li><li data-list="bullet">Longer context windows &amp; the “good-enough” LLM wave</li><li data-list="bullet">Reasoning breakthroughs</li><li data-list="bullet">Vibe coding &amp; new human–model interfaces</li></ol>…and six more major 2025 developments that shaped how we build.The Expert PanelFeaturing practitioners who built, deployed, or evaluated these systems hands-on:<ol><li data-list="bullet"><a href="https://www.linkedin.com/in/nehiljain/" rel="noopener noreferrer" target="_blank">Nehil Jain — ,Member of Technical Staff, Anyscale</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/rohanpra/" rel="noopener noreferrer" target="_blank">Rohan Prasad - Staff Software Engineer, Platforms · EvolutionIQ</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/sophiaskowronski/" rel="noopener noreferrer" target="_blank">Sophia Skowronski — Data Scientist · Breckinridge Capital Advisors</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/matt-squire-a19896125/" rel="noopener noreferrer" target="_blank">Matt Squire — CTO · Fuzzy Labs</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/valdimar-%C3%A1g%C3%BAst-eggertsson-8210236/" rel="noopener noreferrer" target="_blank">Valdimar Eggertsson — AI Team Lead · Smart Data Inc.</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/anaanna417/" rel="noopener noreferrer" target="_blank">Anna Yoon — Tech Lead · Statsig</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/sonamgupta11/" rel="noopener noreferrer" target="_blank">Sonam Gupta — Developer Advocate · Telnyx</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/adamissimo/" rel="noopener noreferrer" target="_blank">Adam Becker — Founder · HeadOn</a></li></ol>Moderated by:<ol><li data-list="bullet"><a href="https://linkedin.com/in/adamissimo/" rel="noopener noreferrer" target="_blank">Arthur Coleman</a><a href="https://linkedin.com/in/adamissimo/" rel="noopener noreferrer" target="_blank"> — CEO, OnlineMatters Inc.</a></li><li data-list="bullet"><a href="https://www.linkedin.com/in/binoy-perera-811282204/" rel="noopener noreferrer" target="_blank">Binoy Perera</a><a href="https://www.linkedin.com/in/binoy-perera-811282204/" rel="noopener noreferrer" target="_blank"> — MLOps Community</a></li></ol>Who This Is ForEngineers, ML practitioners, and technical leaders who want a clear, no-nonsense understanding of what actually moved the field forward this year—and how those shifts will shape your roadmap in 2026.Join us for two hours of high-signal discussion, real-world insight, and informed predictions from the people building the future.📅 Date: December 18th🕚 Time: 11amETJoin the #reading-group channel in the MLOps Community Slack to connect before and after the session.

AI REWIND 2025 - MLOps Reading Group Special

This month's paper: <h1><a href="https://arxiv.org/abs/2506.18203" rel="noopener noreferrer" target="_blank">Shrinking the Generation-Verification Gap with Weak Verifiers</a></h1> Language models are getting better at reasoning but their ability to verify their own outputs still lags behind. This paper tackles that challenge head-on by introducing Weaver, a framework that combines multiple weak verifiers into a single, stronger verifier without relying heavily on labeled data.Weaver uses weak supervision to estimate verifier reliability, normalize inconsistent outputs, and filter low-quality signals, resulting in a unified score that better reflects true response quality. In practice, this approach significantly boosts reasoning and math task performance rivaling models several times larger, such as achieving o3-mini-level accuracy using only Llama 3.3 70B as the generator. <h2>💡 Special Guest - Author of paper:</h2><h2>We’re thrilled to be joined by the Jon Saad-Falcon, Stanford PhD Candidate in Computer Science, to discuss the paper and take questions from the group.</h2> 📅 Date: November 20th🕚 Time: 11amET<h2>Speakers: <a href="https://www.linkedin.com/in/adamissimo/overlay/about-this-profile/" rel="noopener noreferrer" target="_blank">Adam Boaz Becker</a> - Founder, HeadOn. <a href="https://www.linkedin.com/in/anaanna417/overlay/about-this-profile/" rel="noopener noreferrer" target="_blank">Jimin (Anna) Yoon</a> - Tech Lead / Senior Software Engineer</h2><h2>Moderator <a href="https://www.linkedin.com/in/arthurcoleman/overlay/about-this-profile/" rel="noopener noreferrer" target="_blank">Arthur Coleman</a>: CEO, OnlineMatters Inc.</h2> Join the #reading-group channel in the MLOps Community Slack to connect before and after the session.

MLOps Reading Group Nov – Shrinking the Generation-Verification Gap with Weak Verifiers

<h1>The Virtual AI Event That’s Actually... Fun</h1>We know what you’re thinking: another virtual conference. Talking heads, awkward silences, and the constant urge to check your email… we’ve all been there.Not this one.<h3>100% AI Agents in Prod. BUT This Isn't Just Another Zoom Link.</h3>Welcome to Agents in Production: the latest edition of the MLOps × Prosus AI Virtual Conference.We’re bringing together the brightest minds building AI agents, with a high-energy format designed to keep you hooked from start to finish.Last year, companies stopped experimenting with agents and started deploying them in the real world. We heard from the pioneers who turned hype into working systems - and this year, we’re doubling down.Expect real progress, real lessons, and real breakthroughs shaping the future of agentic AI. You’ll get cutting-edge, actionable insights that move you from experimentation to full-scale deployment.<h2>30+ Talks on AI Agents. I Promise You Won’t Log Off Early</h2>Why Attend:<ol><li data-list="bullet">Talks from top experts&nbsp;– Real-world lessons, practical insights, and breakthroughs defining agentic AI.</li><li data-list="bullet">Hilarious skits &amp; live music&nbsp;– Because learning should be fun.</li><li data-list="bullet">High-energy engagement&nbsp;– Interactive moments that make&nbsp;you&nbsp;part of the action.</li></ol>If You Miss This, You’ll Miss:<ol><li data-list="bullet">Hard-won lessons&nbsp;– How leading companies are successfully deploying agents at scale.</li><li data-list="bullet">Deep dives&nbsp;– Technical sessions and workshops from the voices shaping the next generation of AI.</li><li data-list="bullet">Global connections&nbsp;– Network with innovators and practitioners across the ML community.</li></ol>This is your chance to get up to speed on the global AI scene, connect with innovators, and experience a virtual event you’ll&nbsp;actually&nbsp;enjoy.See you there!&nbsp;

Agents in Production - MLOps x Prosus

AI Agent Builder Summit SF

<h1>AI Agents Are Already Working — Let’s Talk About It</h1>Agents are no longer just experiments. From e-commerce to customer support to analytics workflows, they’re quietly getting real work done in production.On July 17, join the MLOps Community for Part 2 of Agents in Production — a virtual event focused on the messy, practical side of building and deploying AI agents.What’s on deck?<ol><li data-list="bullet">Taming complexity: agent memory, behavior control, latency vs. response tradeoffs</li><li data-list="bullet">Stories from the field: How companies are actually using agents in live environments</li><li data-list="bullet">Tooling that works: routing, evaluation, UX, and cost performance in the wild</li></ol><h2>It’s free, it’s global, and it’s going to be packed. </h2>

Agents in Production 2025

The MLOps Community is where machine learning practitioners come together to define and implement MLOps.
Our global community is the default hub for MLOps practitioners to meet other MLOps industry professionals, share their real-world experience and challenges, learn skills and best practices, and collaborate on projects and employment opportunities. We are the world's largest community dedicated to addressing the unique technical and operational challenges of production machine learning systems.

MLOps Community

Home

Forum

Events

Content

People

Contributors

Albums

Clubs

Chat

Channels

Tools

Help

Login to the MLOps Community community to connect with others, attend community events, and more!

Login or Sign Up for even more

To unlock all parts of the community and get the best experience, complete your profile.