MAROKO133 Hot ai: Agentic AI is all about the context — engineering, that is Edisi Jam 05:

📌 MAROKO133 Update ai: Agentic AI is all about the context — engineering, that is

Presented by Elastic

As organizations scramble to enact agentic AI solutions, accessing proprietary data from all the nooks and crannies will be key

By now, most organizations have heard of agentic AI, which are systems that “think” by autonomously gathering tools, data and other sources of information to return an answer. But here’s the rub: reliability and relevance depend on delivering accurate context. In most enterprises, this context is scattered across various unstructured data sources, including documents, emails, business apps, and customer feedback.

As organizations look ahead to 2026, solving this problem will be key to accelerating agentic AI rollouts around the world, says Ken Exner, chief product officer at Elastic.

"People are starting to realize that to do agentic AI correctly, you have to have relevant data," Exner says. "Relevance is critical in the context of agentic AI, because that AI is taking action on your behalf. When people struggle to build AI applications, I can almost guarantee you the problem is relevance.”

Agents everywhere

The struggle could be entering a make-or-break period as organizations scramble for competitive edge or to create new efficiencies. A Deloitte study predicts that by 2026, more than 60% of large enterprises will have deployed agentic AI at scale, marking a major increase from experimental phases to mainstream implementation. And researcher Gartner forecasts that by the end of 2026, 40% of all enterprise applications will incorporate task-specific agents, up from less than 5% in 2025. Adding task specialization capabilities evolves AI assistants into context-aware AI agents.

Enter context engineering

The process for getting the relevant context into agents at the right time is known as context engineering. It not only ensures that an agentic application has the data it needs to provide accurate, in-depth responses, it helps the large language model (LLM) understand what tools it needs to find and use that data, and how to call those APIs.

While there are now open-source standards such as the Model Context Protocol (MCP) that allow LLMs to connect to and communicate with external data, there are few platforms that let organizations build precise AI agents that use your data and combine retrieval, governance, and orchestration in one place, natively.

Elasticsearch has always been a leading platform for the core of context engineering. It recently released a new feature within Elasticsearch called Agent Builder, which simplifies the entire operational lifecycle of agents: development, configuration, execution, customization, and observability.

Agent Builder helps build MCP tools on private data using various techniques, including Elasticsearch Query Language, a piped query language for filtering, transforming, and analyzing data, or workflow modeling. Users can then take various tools and combine them with prompts and an LLM to build an agent.

Agent Builder offers a configurable, out-of-the-box conversational agent that allows you to chat with the data in the index, and it also gives users the ability to build one from scratch using various tools and prompts on top of private data.

"Data is the center of our world at Elastic. We’re trying to make sure that you have the tools you need to put that data to work," Exner explains. "The second you open up Agent Builder, you point it to an index in Elasticsearch, and you can begin chatting with any data you connect this to, any data that’s indexed in Elasticsearch — or from external sources through integrations.”

Context engineering as a discipline

Prompt and context engineering is becoming a discipli. It’s not something you need a computer science degree in, but more classes and best practices will emerge, because there’s an art to it.

"We want to make it very simple to do that," Exner says. "The thing that people will have to figure out is, how do you drive automation with AI? That’s what’s going to drive productivity. The people who are focused on that will see more success."

Beyond that, other context engineering patterns will emerge. The industry has gone from prompt engineering to retrieval-augmented generation, where information is passed to the LLM in a context window, to MCP solutions that help LLMs with tool selection. But it won't stop there.

"Given how fast things are moving, I will guarantee that new patterns will emerge quite quickly," Exner says. "There will still be context engineering, but they’ll be new patterns for how to share data with an LLM, how to get it to be grounded in the right information. And I predict more patterns that make it possible for the LLM to understand private data that it’s not been trained on."

Agent Builder is available now as a tech preview. Get started with an Elastic Cloud Trial, and check out the documentation for Agent Builder here.

Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact [email protected].

🔗 Sumber: venturebeat.com

📌 MAROKO133 Update ai: Vibe coding platform Cursor releases first in-house LLM, Co

The vibe coding tool Cursor, from startup Anysphere, has introduced Composer, its first in-house, proprietary coding large language model (LLM) as part of its Cursor 2.0 platform update.

Composer is designed to execute coding tasks quickly and accurately in production-scale environments, representing a new step in AI-assisted programming. It's already being used by Cursor’s own engineering staff in day-to-day development — indicating maturity and stability.

According to Cursor, Composer completes most interactions in less than 30 seconds while maintaining a high level of reasoning ability across large and complex codebases.

The model is described as four times faster than similarly intelligent systems and is trained for “agentic” workflows—where autonomous coding agents plan, write, test, and review code collaboratively.

Previously, Cursor supported "vibe coding" — using AI to write or complete code based on natural language instructions from a user, even someone untrained in development — atop other leading proprietary LLMs from the likes of OpenAI, Anthropic, Google, and xAI. These options are still available to users.

Benchmark Results

Composer’s capabilities are benchmarked using "Cursor Bench," an internal evaluation suite derived from real developer agent requests. The benchmark measures not just correctness, but also the model’s adherence to existing abstractions, style conventions, and engineering practices.

On this benchmark, Composer achieves frontier-level coding intelligence while generating at 250 tokens per second — about twice as fast as leading fast-inference models and four times faster than comparable frontier systems.

Cursor’s published comparison groups models into several categories: “Best Open” (e.g., Qwen Coder, GLM 4.6), “Fast Frontier” (Haiku 4.5, Gemini Flash 2.5), “Frontier 7/2025” (the strongest model available midyear), and “Best Frontier” (including GPT-5 and Claude Sonnet 4.5). Composer matches the intelligence of mid-frontier systems while delivering the highest recorded generation speed among all tested classes.

A Model Built with Reinforcement Learning and Mixture-of-Experts Architecture

Research scientist Sasha Rush of Cursor provided insight into the model’s development in posts on the social network X, describing Composer as a reinforcement-learned (RL) mixture-of-experts (MoE) model:

“We used RL to train a big MoE model to be really good at real-world coding, and also very fast.”

Rush explained that the team co-designed both Composer and the Cursor environment to allow the model to operate efficiently at production scale:

“Unlike other ML systems, you can’t abstract much from the full-scale system. We co-designed this project and Cursor together in order to allow running the agent at the necessary scale.”

Composer was trained on real software engineering tasks rather than static datasets. During training, the model operated inside full codebases using a suite of production tools—including file editing, semantic search, and terminal commands—to solve complex engineering problems. Each training iteration involved solving a concrete challenge, such as producing a code edit, drafting a plan, or generating a targeted explanation.

The reinforcement loop optimized both correctness and efficiency. Composer learned to make effective tool choices, use parallelism, and avoid unnecessary or speculative responses. Over time, the model developed emergent behaviors such as running unit tests, fixing linter errors, and performing multi-step code searches autonomously.

This design enables Composer to work within the same runtime context as the end-user, making it more aligned with real-world coding conditions—handling version control, dependency management, and iterative testing.

From Prototype to Production

Composer’s development followed an earlier internal prototype known as Cheetah, which Cursor used to explore low-latency inference for coding tasks.

“Cheetah was the v0 of this model primarily to test speed,” Rush said on X. “Our metrics say it [Composer] is the same speed, but much, much smarter.”

Cheetah’s success at reducing latency helped Cursor identify speed as a key factor in developer trust and usability.

Composer maintains that responsiveness while significantly improving reasoning and task generalization.

Developers who used Cheetah during early testing noted that its speed changed how they worked. One user commented that it was “so fast that I can stay in the loop when working with it.”

Composer retains that speed but extends capability to multi-step coding, refactoring, and testing tasks.

Integration with Cursor 2.0

Composer is fully integrated into Cursor 2.0, a major update to the company’s agentic development environment.

The platform introduces a multi-agent interface, allowing up to eight agents to run in parallel, each in an isolated workspace using git worktrees or remote machines.

Within this system, Composer can serve as one or more of those agents, performing tasks independently or collaboratively. Developers can compare multiple results from concurrent agent runs and select the best output.

Cursor 2.0 also includes supporting features that enhance Composer’s effectiveness:

• In-Editor Browser (GA) – enables agents to run and test their code directly inside the IDE, forwarding DOM information to the model.

• Improved Code Review – aggregates diffs across multiple files for faster inspection of model-generated changes.

• Sandboxed Terminals (GA) – isolate agent-run shell commands for secure local execution.

• Voice Mode – adds speech-to-text controls for initiating or managing agent sessions.

While these platform updates expand the overall Cursor experience, Composer is positioned as the technical core enabling fast, reliable agentic coding.

Infrastructure and Training Systems

To train Composer at scale, Cursor built a custom reinforcement learning infrastructure combining PyTorch and Ray for asynchronous training across thousands of NVIDIA GPUs.

The team developed specialized MXFP8 MoE kernels and hybrid sharded data parallelism, enabling large-scale model updates with minimal communication overhead.

This configuration allows Cursor to train models natively at low precision without requiring post-training quantization, improving both inference speed and efficiency.

Composer’s training relied on hundreds of thousands of concurrent sandboxed environments—each a self-contained coding workspace—running in the cloud. The company adapted its Background Agents infrastructure to schedule these virtual machines dynamically, supporting the bursty nature of large RL runs.

Enterprise Use

Composer’s performance improvements are supported by infrastructure-level changes across Cursor’s code intelligence stack.

The company has optimized its Language Server Protocols (LSPs) for faster diagnostics and navigation, especially in Python and TypeScript projects. These changes reduce latency when Composer interacts with large repositories or generates multi-file updates.

Enterprise users gain administrative control over Composer and other agents through team rules, audit logs, and sandbox enforcement. Cursor’s Teams and Enterprise tier…

Konten dipersingkat otomatis.

🔗 Sumber: venturebeat.com