Databricks

As AI teams adopt more autonomous agents, one challenge becomes clear: they don't naturally work together. Omnigent is an open-source meta-harness that sits above individual agent harnesses, routing work across multiple agents through a single orchestration layer. Instead of managing agents in isolation, teams can compose different models and agent harnesses into unified workflows, with shared governance and collaboration built in. Explore the project on GitHub: github.com/omnigent-ai/om…

Omnigent 0.5.0 is Released! Highlights in this release: 📱 Omnigent for iOS apps.apple.com/us/app/omnigen… 🧩 Generic ACP harness 🔎 Command palette & richer search 🎨 More appearance & customization setting 💬 Message steering & queuing ☸️ Kubernetes deployment 🌐 Embedded browser Enjoy! Read more about this release: omnigent.ai/releases/0.5.0

@omnigent_ai Excited for this!

Omnigent Deep Dive + Patio Social 🌤️ Join us for a midday technical talk on the architecture, governance, and real-world deployment of the open-source meta-harness for AI agents, then mingle over food and drinks on the @Databricks Seattle patio! 📅 Wed, July 29 🕛 12:00–1:30 PM PT 🔗 RSVP: luma.com/tech_workshop_… #Omnigent #SeattleTechWeek

OpenAI GPT-5.6 Sol, GPT-5.6 Terra, and GPT-5.6 Luna are now available as Databricks-hosted models through Model Serving. Built for complex problem-solving, coding, and budget-friendly workloads, the GPT-5.6 series uses fewer tokens to complete tasks, making the models highly efficient and cost-effective. Access all three through Foundation Model APIs on a pay-per-token basis. #openai-gpt-56-sol-gpt-56-terra-and-gpt-56-luna-now-available-as-databricks-hosted-models" target="_blank" rel="nofollow noopener">docs.databricks.com/aws/en/release…

"Every stateful system has got to have the ability to create a staging environment, branch what you have, and give agents permission to make mistakes." Databricks' VP @nikitabase joins 𝘐𝘵'𝘴 𝘈𝘣𝘰𝘶𝘵 𝘋𝘢𝘵𝘢 to discuss why branching and version control are becoming foundational for AI-native software development, how teams can manage schema drift as AI accelerates development, and the architectural ideas behind Lakebase. Watch the full conversation: youtube.com/watch?v=ZB2sVr…

Genie One is the data-smart AI coworker that helps business users move from insight to action. With Genie Agents, teams can turn prompts into shareable, autonomous agents that reason over structured and unstructured data. Genie Ontology provides the context layer that helps those agents identify trustworthy sources and understand enterprise knowledge. Together, they deliver AI that's grounded in business context, with higher accuracy and performance. databricks.com/blog/introduci…

Gartner’s Magic Quadrant for Analytics and Business Intelligence (BI) is out, and Databricks was named a Visionary in our first appearance, the highest debut for any vendor in this MQ’s 20+ year history. BI has already changed. Anyone can drill from a signal down to the truth behind it and agentic loops make sure every answer has the full story. That’s where we're ahead with Genie and AI/BI. I use it every day to see the key signals, understand what changed and why, and decide what to do about it. Huge congratulations to the teams, and thank you to our customers.

The best of #DataAISummit is now available on demand. Watch keynotes and select sessions featuring the latest product announcements, technical deep dives, and expert conversations. Start watching: databricks.com/dataaisummit?u…

“The attackers are now hyper-automated with agents and large language models that can find vulnerabilities that haven't yet been found by the people who wrote the software.” Databricks’ @krioukov and NAB’s Patrick Wright joined @theCUBE to discuss what security looks like in the age of AI. Their conversation covers how broader data context, automation, and agentic workflows can help security teams move beyond manual investigations toward machine-speed defense. youtube.com/watch?v=MBx51U…

At 11k employees, our AI costs are going up. Which model & harness should we use to lower cost but also retain great quality? We didn't want to blindly trust public benchmarks. So we ran a comprehensive evaluation on our tasks, code base, infra. It's been produced by more than 3,000 software engineers, spans 3 hyperscalar clouds and many languages and tasks. The results are surprising. We find that for the SAME mdoel, the choice of harness can significantly save costs (~2x). We also find that GLM 5.2 performs extremely well. We run Omnigent in front of these and can easily multiplex different harnesses and models for different tasks. Check it out: databricks.com/blog/benchmark…

We benchmarked coding agents on real engineering tasks across Databricks' multi-million line codebase to better understand the relationship between cost and performance. The benchmark surfaced several key findings. Top-tier performance now comes from a mix of proprietary and open models, token pricing is a poor proxy for real-world cost, and harness choice can dramatically impact both cost and quality. Read how we built the benchmark and what we learned → databricks.com/blog/benchmark…

We benchmarked coding agents on our own internal tasks at Databricks and learned a lot! There are many surprising opportunities to lower cost and increase quality, and many models including open source ones are truly competitive now. 🧵

Agent Bricks has expanded into a comprehensive agent platform for developers. The latest capabilities are designed to help solve three critical challenges in production AI agents: 1️⃣ Choice 2️⃣ Context 3️⃣ Control New capabilities include broader model support, managed agent memory, MCP connectivity, secure sandboxes, and Unity AI Gateway for governance, monitoring, and cost controls, helping developers focus on building production AI agents. databricks.com/blog/agent-bri…

Happy 30th birthday Postgres! 🎉 What an accomplishment, almost nothing in tech has stayed relevant for 30 years. From humble beginnings as a Berkeley research project to becoming the default for everything from side projects to massive production systems. 30 years later, the database and the community have more momentum than ever. It's why we built Neon on Postgres, serverless, but 100% the Postgres you love. 🐘

Meet Genie, your data-smart AI coworker. Generic AI assistants don't know your business. Genie does. Grounded in your business context, Genie helps every team find trusted insights and turn answers into action. databricks.com/genie?utm_sour…

The next evolution of Lakeflow brings ingestion, transformation, and orchestration together in a unified platform, fully integrated and centrally governed by Unity Catalog. New capabilities include Lakeflow Designer, Genie ZeroOps, 100+ Lakeflow Connect connectors, expanded Zerobus Ingest interfaces, Real-Time Mode for Spark Declarative Pipelines, and enhanced Lakeflow Jobs orchestration. A unified, real-time data foundation for agentic AI. databricks.com/blog/lakeflow-…

Omnigent now supports contextual policies, enabling session-aware governance for AI agents. Instead of evaluating actions one at a time, contextual policies track what has happened throughout a session to determine whether the next action should proceed. That enables richer controls such as per-session budgets, dynamic risk scoring, and least-privilege access. As a meta-harness, Omnigent applies these policies across supported agent harnesses, including Claude Code, Codex, and custom agent frameworks. Learn more: databricks.com/blog/contextua…

Most coding agents have very limited features for controlling security: just a list of actions to allow, deny, or ask users on. In @omnigent_ai, we introduced Contextual Policies, which can track *state* about a session to manage both security and cost in a far more flexible way.

🧞 Genie Code is expanding to support more complex, agentic data and ML work on Databricks. New capabilities include: → A full-page command center to manage complex, multi-threaded work with thread status, review points, and quick access to instructions, skills, and connectors. → Native intelligence across MLflow, Model Serving, and compute, turning the same agent you already use into a specialist for production ML engineering. Coming soon, scheduled tasks will let Genie Code run work autonomously and hand back the results for review. See it in action: databricks.com/blog/whats-new…

Data + AI Summit 2026 marked another milestone in the @OpenAI and Databricks partnership. Keynotes, executive conversations, customer stories, a hackathon, and the Grounded Reasoning Cup showcased how OpenAI's frontier intelligence and Databricks' enterprise context and governance are helping organizations build production-ready AI. Explore the biggest moments from the summit, real-world customer deployments, and what's next for the partnership. databricks.com/blog/openai-an…