Software Engineering AI

Cognition's Scott Wu says AI coding agents shouldn't replace humans. Cognition, known for its AI coding agent Devin, recently raised $1 billion at a $26 billion valuation. Wu emphasized that the goal is not to replace human programmers but to assist them. Wu, a former competitive programmer, believes AI should augment human work, allowing programmers to focus on creative tasks while AI handles maintenance. He envisions AI agents entering various fields, always with humans deciding their role. Read more: https://lnkd.in/guR4iPPJ 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

Visa has invested in and partnered with Replit to integrate Visa Intelligent Commerce into Replit's platform. This collaboration aims to help developers build applications that can initiate secure transactions using Visa's global network. Replit has also launched self-serve enterprise access, allowing organizations to purchase enterprise-grade compliance and controls directly. Replit's new Solution Partner Program includes partners like Accenture and Slalom to support enterprise deployment at scale. Read more: https://lnkd.in/gqwGCEZY 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

Replit has launched the Replit Solution Partner Program. This new program connects enterprises with a network of expert partners, including Accenture, Slalom, and Hexaware, to help adopt and scale AI software creation. Replit's Chief Revenue Officer, Ghazi Masood, said the program provides enterprises with a trusted network to deploy and scale Replit across their organizations. The program follows the launch of Agent 4, Replit's latest offering, and expands its ecosystem with service partners for enterprise deployment. Read more: https://lnkd.in/gFBTVwSd 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

AHEAD and Cursor have announced a partnership to enhance AI-native engineering in enterprises. The collaboration aims to scale AI-native engineering across teams and workflows, moving beyond individual developer adoption. Cursor's platform will be integrated with AHEAD's expertise in enterprise advisory and change management to transform software delivery processes. Eric Kaplan from AHEAD said the partnership is about helping clients transition from adoption to transformation in the AI era. Read more: https://lnkd.in/ghW25-Zd 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

Cognition, an AI coding startup, has raised over $1 billion at a $25 billion pre-money valuation. The funding round was led by Lux Capital, General Catalyst, and 8VC, with participation from existing and new investors. Cognition's AI software engineer, Devin, has seen significant growth, with enterprise usage increasing by over 10 times this year. The company reports a $492 million annualized revenue run-rate, with major clients including Mercedes-Benz AG, NASA - National Aeronautics and Space Administration, and Goldman Sachs. Read more: https://lnkd.in/g8XaXWfZ 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

OpenAI generated about $5.7 billion in revenue in the first quarter of 2026, surpassing Anthropic by nearly $1 billion. The growth was driven by OpenAI's coding agent Codex, business sales, and testing ads on ChatGPT. Despite this, Anthropic's recent growth rate suggests it may soon surpass OpenAI. Anthropic's annualized revenue recently approached $45 billion, while OpenAI's was around $30 billion. Anthropic expects its Q2 revenue to double to $11 billion with a $600 million profit. OpenAI's first-quarter results helped secure $122 billion in new funding, while Anthropic is raising $30 billion. Both companies are preparing for potential IPOs. Read more: https://lnkd.in/gupXy27H 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

Salesforce plans to invest $300 million in Anthropic's AI tokens. CEO Marc Benioff said the investment aims to enhance AI coding tools, which have already increased productivity by over 30%. Salesforce has paused hiring new software engineers, attributing this to productivity gains from AI, but plans to hire more salespeople to promote AI products. Benioff emphasized that AI tools are transforming engineering roles rather than replacing human developers. Read more: https://lnkd.in/g42vWZut 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

Trust3 AI has launched MCP Security to enhance the governance and security of enterprise agentic AI workloads. This new solution is part of Trust3 AI's enterprise agent control plane, providing a unified trust layer for connecting AI agents with business data and systems securely. The MCP Security addresses risks associated with autonomous AI architectures by offering robust identity access management and a comprehensive Agent DOS platform. Don Bosco, Co-Founder of Trust3 AI, said that security must be built into the protocol itself to ensure mission-critical workloads can be run with confidence. Read more: https://lnkd.in/gZtqsapc 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

1Password has teamed up with OpenAI to prevent AI coding agents from leaking credentials. The collaboration introduces a just-in-time credential model for OpenAI Codex, ensuring credentials are not exposed in prompts, code repositories, or model contexts. 1Password's CTO, Nancy Wang, said that persistent credentials are already compromised, highlighting the importance of just-in-time credentials for AI-native development. Read more: https://lnkd.in/gwsHJsgR 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews

Bugcrowd has launched Reinforcement Learning Environments to enhance AI models' real-world security skills. The new offering helps AI developers train models to find, exploit, and fix software vulnerabilities using real software rather than synthetic data. Dave Gerry, CEO of Bugcrowd, said the environments allow AI to learn from real vulnerabilities, bridging the gap between training and real-world application. The platform includes numerous training environments built from open-source vulnerabilities, providing immediate feedback to improve AI models. Read more: https://lnkd.in/g6acC_3g 📰 Subscribe to the weekly AI Programming Weekly: https://lnkd.in/eUQ-KTc2 #ai #artificialintelligence #ainews