Thoughtworks technologist Birgitta Böckeler tested running local large language models for agentic coding on Apple M3 Max and M5 Pro machines, finding that while RAM constraints limit model size to 15…
A Bitcoin Policy Institute report found that a nonprofit network linked to the Chinese Communist Party has blocked or delayed $23.6 billion in AI data center projects across 14 U.S. states, including …
Cassie Shum, product engineering lead at RelationalAI, presented on GraphRAG at a conference, explaining how knowledge graphs enhance retrieval workflows by adding semantic structure. She contrasted t…
A developer argues that green test suites only measure conformance to a specification, not whether a change actually improves the system. The post contrasts Spec-Driven Development, which stops at cor…
Developers building AI agents face reliability challenges in production, including non-determinism, token bloat, and cascading failures. To address this, engineering teams should adopt deterministic w…
Bayer AG and Thoughtworks developed the Preclinical Information Center (PRINCE), a cloud-hosted platform using Agentic Retrieval-Augmented Generation and Text-to-SQL to integrate decades of safety stu…
Zed 1.7.2, released June 17, introduces automatic context compaction for AI agent sessions, which summarizes earlier conversation when the context window reaches 90% capacity to prevent performance de…
Enterprise architecture governance is evolving toward 'architecture-as-code,' where architectural standards and constraints become machine-readable artifacts that can be versioned, tested, and continu…
A debate is emerging in engineering circles that AI has killed Agile and brought back Waterfall, but the reality is more nuanced: the high cost of writing code that made Agile necessary is gone, repla…
OpenAI transformed ChatGPT into an enterprise-focused superapp with coding tools and AI agents, ending the chatbot-as-product era. New research quantified token usage in AI coding workflows, revealing…
Sarang Kulkarni of Thoughtworks presented at the Arc of AI Conference 2026 on building deep research AI agents for healthcare and pharmaceutical R&D, detailing the development of an "Agentic RAG++" sy…
The article argues that if an organization mandates using LLMs to minimize coding time, programmers may stop reading LLM-generated code, treating it like machine code, and instead shift rigor to speci…
Based on the article by Birgitta Böckeler, a software harness contains two distinct components: "guides," which proactively shape an agent's behavior before it produces output, and "sensors," which re…