IBM pronounces self-hosted deployment for IBM Bob
IBM Bob, the corporate’s agentic software program growth platform, helps groups transfer past code era to use AI to the work of software program supply and modernization. The platform permits organizations to convey AI software program growth to on-premises, private-cloud, sovereign-cloud, and air-gapped environments the place they’ve management, the corporate mentioned.
IBM Bob’s self-hosted deployment possibility brings AI to customer-managed infrastructure and is particularly designed for organizations that require higher management over information residency, safety insurance policies, AI governance, and software program growth workflows. The self-hosted model of IBM Bob permits enterprises to deploy AI software program growth and modernization capabilities in environments the place their code, utility context, and information already reside.
“The way forward for enterprise AI will depend upon safety, governance and sovereignty,” mentioned Neel Sundaresan, GM of AI and Automation, IBM. “Organizations want AI that operates inside environments they’ve management over, particularly when working with delicate code and controlled information. Bob’s self-hosted deployment gives a means for enterprises to convey agentic AI on to these environments to allow them to profit from the expertise whereas sustaining safety, compliance, and operational management.” The platform brings AI to the info as an alternative of transferring the info to the AI.
For added details about self-hosted choices for IBM Bob, go to: https://www.ibm.com/new/bulletins/ibm-bob-expands-to-self-hosted-environments-for-sensitive-and-mission-critical-enterprise-software
A New Agentic Expertise: JetBrains Air in IDEs
JetBrains Air, an open system of merchandise for agentic growth,in the present day is launching the early entry program for Air within the 2026.3 EAP builds of JetBrains IDEs, or as a plugin on the JetBrains Market.
Air works with the brokers you already use, reminiscent of Codex, GitHub Copilot, Junie, Cursor, and different ACP-compatible brokers. It detects brokers already put in in your machine and makes it simple to convey them instantly into the IDE. With JetBrains Air in IDEs it’s also possible to use your Claude subscription in Air’s terminal and full-screen tab.
You don’t want a JetBrains AI subscription to make use of your brokers with Air. Use your current AI supplier subscriptions or JetBrains AI if you have already got it.
Constructed-in IDE expertise and instruments give supported brokers entry to specialised JetBrains experience and workflows for duties like debugging, efficiency profiling, database exploration, and semantic code search. This will help brokers produce higher outcomes and, for some duties, use fewer tokens.
Improvement doesn’t finish when code is generated – it nonetheless must be run, reviewed, and validated. That’s precisely what an IDE is constructed for. It provides you the instruments to grasp and assessment the agent’s work – from diffs and navigation to inspections and extra.
Constructed-in IDE expertise and instruments give supported brokers entry to specialised JetBrains experience and workflows for duties like debugging, efficiency profiling, database exploration, and semantic code search.
Qodo 3.0 Brings High quality Management to Agentic Software program Improvement
With this launch, Qodo expands its high quality infrastructure to enterprises constructing agentic software program factories, robotically connecting agent workflows to enterprise codebase data, architectural requirements, and governance necessities.
Many organizations are struggling to take care of high quality and governance on the tempo brokers now produce code. In keeping with a current trade research, 91% of engineering leaders worry being liable to shedding management of their codebase, with governance and visibility cited as the highest danger at virtually 30%. Qodo 3.0 addresses this by injecting qc all through the agentic SDLC, guided by Qodo’s Knowledge Base that gives brokers with a repeatedly up to date understanding of a corporation’s codebase, requirements, structure, and pull request historical past.
With Qodo 3.0, high quality is embedded into code era earlier than a pull request exists by surfacing a corporation’s coding requirements into an agent’s workspace and working Qodo’s assessment engine on code earlier than it’s dedicated. Governance and codebase high quality may be monitored and risk-assessed with the Qodo Software program Map, which robotically maps a corporation’s repositories, calculates the blast radius of a proposed change, and surfaces high quality points as a warmth map throughout the structure.
