Akka unveils Agentic AI Platform to boost enterprise adoption
Akka has launched its Agentic AI Platform, which aims to help enterprises accelerate the building, deployment, and evaluation of agentic systems while reducing the required computational resources.
The platform introduces a suite of integrated solutions: Akka Orchestration, Akka Agents, Akka Memory, and Akka Streaming. These offerings are designed to address common challenges faced by IT leaders as they move agentic AI projects from pilot stages into full-scale production.
Managing complexity
According to Akka, the complexity introduced by agentic AI, such as emergent behaviours and non-deterministic decision-making, can create significant hurdles for enterprise adoption. The company outlines that these factors limit the ability of IT departments to trust, scale, and reliably operate AI systems for critical business applications, as traditional deterministic approaches are no longer sufficient.
"Agentic systems are forcing IT leaders to rethink their technology stack," said Tyler Jewell, Akka's CEO. "IT systems must adapt from controlling predefined workflows to managing intelligent, adaptive systems operating in open-ended environments that include non-deterministic LLMs. Scaling these systems and providing dependable outputs is a tremendous challenge and redefines the meaning of an SLA. Akka is unique in that we're bringing IT the tools to solve this issue at enterprise scale, with enterprise confidence."
The company positions the Akka Agentic Platform as a comprehensive set of tools that allows enterprises to manage the orchestration, memory, and streaming requirements of agentic AI with greater reliability.
Platform components
The platform's four major capabilities each address different operational needs. Akka Orchestration is focused on managing multi-agent systems under a range of conditions, including infrastructure failures and workflow interruptions. It allows for governance of agents, tools, APIs, and resources, and supports a variety of workflow patterns, including human-in-the-loop operations.
Akka Agents lets enterprises develop autonomous and goal-directed software agents that can interact with third-party systems using common protocols. Akka Memory provides an in-memory, sharded data architecture designed to deliver high-speed data access and storage for agent context and personalisation, with capabilities for both short- and long-term retention. Akka Streaming offers continuous data processing for ambient, adaptive, and real-time AI scenarios, enabling the ingestion, aggregation, and enhancement of live streams from varied sources, including IoT devices and multimedia content.
Akka states that these solutions are aimed at giving IT leaders greater certainty about meeting operational service-level agreements (SLAs) relating to system availability, recovery, safety, and output accuracy. The company also highlights the platform's ability to support a range of use cases, including intelligent automation, autonomous and adaptive AI, analytical systems, and digital twins, across diverse sectors.
Customer feedback
The utility of the platform is reflected in testimonials from enterprises that have already deployed Akka's technologies. Tubi, a video-streaming service owned by Fox Corporation, commented, "Akka has enabled Tubi to provide customer experiences unlike any other in the video-on-demand space."
Swiggy, a company that recently completed a public listing, cited benefits for its machine learning operations. The company stated it achieved a "2x latency improvement in Swiggy ML and AI platform."
Llaama, a healthcare AI startup, referenced resource savings and operational efficiencies: "What Akka gave us out-of-the-box would have taken months of DevOps and database administration to achieve ourselves."
Availability
All new solutions are offered as part of the current Akka licence and are available for enterprise users. The company reports that its framework is already in use by established brands and emerging startups for distributed systems serving millions of users.
