CRAFT (Create, Remember, Assemble, Fine-tune, Trust) empowers business leaders – not just developers – to build and deploy intelligent agent systems using plain English commands
Emergence AI, an agentic frontier company, announced the launch of CRAFT, the world’s first natural language, self-serve platform designed to modernize one of enterprise IT’s most costly and persistent challenges: the outdated, labor-intensive data pipeline – an area representing over $200 billion in global spend. CRAFT empowers enterprise users – from analysts to executives, regardless of technical background – to turn business goals into intelligent, self-verifying, multi-agent systems – fast. The result: faster understanding of key metrics, greater productivity, and a radical shift in how enterprises execute data-driven work.
What once took weeks or months of effort by teams of engineers – manually coding ETLs, migrating schemas, cleaning datasets, stitching together tools, and building ML models to bridge gaps across the data pipeline in order to extract actionable insights – can now happen in minutes. CRAFT empowers business users to describe their goals in plain English – and deploy a swarm of specialized agents to automatically build, test, and run the necessary workflows. CRAFT’s autonomous agents operate across the entire data pipeline, reducing the need for deep technical skills and filling critical gaps caused by the global shortage of data scientists and engineers. A recent report from the United States Data Science Institute highlighted that the demand for data scientists in 2026 will be over by 50% more than the supply in the US.
While CRAFT goes far beyond basic RPA-style automations, it doesn’t dismiss them – in fact, it makes them radically easier. If you’re automating across multiple enterprise systems to build RPA-style workflows, CRAFT’s Agents Creating Agents (ACA) system enables you to do so effortlessly, without writing a single line of code. And for teams ready to go further, CRAFT also serves as a launchpad for more intelligent, adaptive, agentic systems – capable of reasoning, self-improvement, and domain-specific execution. In doing so, CRAFT bridges the gap between task automation and truly autonomous enterprise workflows.
Working with design partners across sectors like semiconductors, oil and gas, healthcare, telecom, and financial services, CRAFT is already enabling teams to dramatically reduce manual work, accelerate analytics, and deliver outcomes at the speed of business.
Also Read: DeepL Adds New Languages for Text and Document Translation
“CRAFT is a launchpad for enterprises to build truly agentic systems,” said Satya Nitta, Co-Founder and CEO of Emergence AI. “Using just plain English, you get dynamic, self-verifying agent workflows – no code, no drag-and-drop templates, just outcomes. But that’s only the beginning. CRAFT also introduces advanced planning and reasoning, self-improvement, and long-term memory modules – pushing the boundaries of what intelligent systems can do. It’s a transformative shift that accelerates time to value and moves enterprises from reactive analytics to continuous, AI-led execution.”
“CRAFT brings the intelligence and flexibility needed to compress weeks of work into minutes,” added Nitta. “Enterprises can now handle machine-scale data faster, more efficiently, and at significantly lower cost – achieving productivity gains that once required entire teams. Furthermore, this approach transforms billable hours into reusable software – bending the enterprise cost curve and freeing integrators to pursue higher-margin transformation work.”
“CRAFT also marks a new category in enterprise automation: Agent Crafting,” Nitta said. “This isn’t just about building agents – it’s full lifecycle management. Purpose-built agent swarms can now code, test, orchestrate, and self-govern – replacing brittle scripts with intelligent systems that adapt in real time.”
Emergence is partnering with NI/Emerson to bring CRAFT driven data analysis on hundreds of gigabytes of data to improve chip yield for semiconductor companies.
Eran Rousseau, VP & Fellow of Business Development and Strategic Partnerships of NI/Emerson, said, “Previously, by the time manual analysis was complete, the next production batch was underway, limiting timely action. With CRAFT, we gain real-time insights aligned with complex test workflows, enabling smart automation to improve yields and detect issues quickly. Integrated with NI’s O+ solution, CRAFT AI generates insights that translate into immediate actions, allowing customers to focus on solving critical challenges rather than on lengthy reports. This represents the future of semiconductor operations.”
Intuitive Interface for Powerful Multi-Agent Workflows
At the heart of CRAFT is a user interface that blends intuitive design with technical depth – setting it apart from static dashboards and rigid templates offered by many AI enterprise platforms. Users can create projects, define workflows, and generate fully functional agents using natural language – no coding or canned drag and drop blocks required. CRAFT’s AI agents interpret natural language prompts, analyze existing codebases, and generate high-quality, context-aware agents tailored to specific requirements.
Powering CRAFT’s unique approach is Agents Creating Agents (ACA), a novel system Emergence demonstrated earlier this year, where agents dynamically build and improve other agents in real time, enabling complex multi-agent workflows with minimal human input. By enabling the creation of agent-based tools and connectors, the platform compresses design time – accelerating deployment, streamlining cross-system data flows, and enabling faster automation and insights across fragmented enterprise systems – all with less overhead.
Inside CRAFT:
- Create – Quickly build with native or custom agents using natural language – no coding required. Integrate with your tools, data, and LLMs to deploy agents directly from conversation.
- Remember – CRAFT retains memory across sessions and workflows, enabling agents to learn, adapt, and stay context-aware over time.
- Assemble – Design workflows manually or let agents plan and execute dynamically. Leverage the Agent Registry – a central directory of all available agents (CRAFT-native, enterprise-created, and third-party) – to reuse and scale automation across projects. Includes starter kits, multi-agent support, and live/offline execution views.
- Fine-Tune – Customize agent behavior, edit code, or let agents self-improve via feedback and self-play – blending no-code simplicity with developer control.
- Trust – Deploy flexibly across on-prem, VPC, or cloud-native environments with full control over privacy, scaling, and infrastructure. Enterprise-grade governance with SOC 2, GDPR, audit logging, role-based access, and runtime safeguards. Full transparency, privacy, and human-in-the-loop oversight built in.
“Enterprise constraints – regulatory, structural, procedural – are no longer blockers for adopting AI; they’re scaffolds for building more intelligent agents,” added Nitta. “With CRAFT, these constraints become the backbone of agents that learn faster, verify better, and adapt smarter.”
Source: Businesswire
