AI Coding Assistants

Today's Stories

1. 1

   IBM Releases Granite 4.2 Reasoning LLMs

   IBM released Granite 4.2, its first family of reasoning large language models (LLMs, AI systems that understand and generate text), in three sizes: 3B, 8B, and 30B. The models are pre-trained from scratch on roughly 15T tokens and extend the context window to 512K tokens. The 8B and 30B models add agentic training, meaning they can operate as agents that use tools, edit and run code, and search the web. All models include a thinking/non-thinking switch and native tool calling, making them practical for business automation and complex tasks.

   The models are released under the Apache 2.0 license, and the larger models (8B and 30B) can act as agents inside real sandboxed environments, which could enable more autonomous AI workflows.

2. 2

   Andrew Ng relaunches DeepLearning.AI around AI engineering

   Andrew Ng, cofounder of Google Brain and Coursera, is relaunching DeepLearning.AI with a focus on AI engineering. The direction is based on analysis of over 10,000 job postings and dozens of interviews with AI experts and hiring managers. Ng's framework highlights four key skills—building and deploying AI applications, software engineering fundamentals, using coding agents, and shaping the build. The post argues these skills matter beyond just AI engineer job titles, and the commentary notes that LLMs raise the ceiling for skilled developers more than they raise the floor for novices.

   The skill of using coding agents is evolving quickly and now includes orchestrating multiple agents and avoiding pitfalls like agent messing up production databases. The post also highlights the importance of product sense and business context in shaping the build, a skill that wasn't foreseen in the original AI engineer essay.

3. 3

   IMDEX unifies geological data platform in 8 months with Cursor

   Australian mining technology company IMDEX used the AI coding tool Cursor to unify two flagship geological mapping tools into a single platform in eight months. Without Cursor, the data platform work alone would have required 20 additional engineers, two to three years, and millions of dollars. The migration of a legacy Angular app to a React micro-frontend was expected to take another three to five years. For Japanese resource-related companies, the shorter development cycles enabled by AI coding tools could also accelerate business transformation.

   IMDEX is rolling out Cursor across the company and has already built a new customer insights app in under 30 days and a sales training tool in just under two weeks using Cursor.

4. 4

   Devin Boeing teaches tech to close digital gap

   Devin Boeing, 26, founded Cryptic Institute to teach technology and finance to everyday people. He discussed this on MSR On the Radar with host Aaliyah Demry. Boeing started the institute after learning Minnesota ranked 49th in digital literacy access for underserved communities. His free four-week classes, held with partners like the Minneapolis Urban League, aim to help youth and elders navigate today's digital world.

   Cryptic Institute partners with local community centers, libraries, and the Minneapolis Urban League. Classes cover basic tools like Microsoft Word to newer topics like AI and cryptocurrency.

5. 5

   GPT-5.6 now in Kiro

   OpenAI's GPT-5.6 is now available in Kiro. It helps developers plan, build, review, and test software with better price-performance.

   Developers can use it in Kiro for software development tasks.

6. 6

   Canonical backs AI-driven C-to-Rust translation PhD

   Canonical is co-funding a three-year PhD project at the University of Bristol to investigate using LLMs to translate large C codebases into safe, behaviorally correct, and maintainable Rust. The project will examine snap-confine and AppArmor. The effort aims to preserve knowledge embedded in mature codebases, which may be lost in clean human rewrites. Traditional source-to-source translators often produce Rust that still relies on unsafe operations and awkward C idioms.

   The project is co-sponsored with UK Research and Innovation. Seager will oversee it alongside Professor Meng Wang and Dr Cristina David. Results may inform whether AI can handle codebases comprising hundreds of thousands of lines of C.