5 Quick Links For Devs: Week 3, 2026

January 13, 2026 · Jacob E. Dawson

Don't fall into the anti-AI hype

The founder of Redis, and extremely well-respected coder Antirez wants you to know that things are changing:

"I have a single suggestion for you, my friend. Whatever you believe about what the Right Thing should be, you can't control it by refusing what is happening right now. Skipping AI is not going to help you or your career."

I totally agree here - right now there are 3 camps of experienced coders: 1.) Those that understand we're in a new paradigm and are learning how to use the new tools as quickly as possible 2.) Those with their heads in the sand trying to pretend that LLMs are useless and "only hallucinate" and 3.) Those that don't have an opinion either way because they haven't been paying attention.

Try to put yourself in camp 1.

Opus 4.5 is going to change everything

"If you had asked me three months ago about these statements, I would have said only someone who’s never built anything non-trivial would believe they’re true. Great for augmenting a developer’s existing workflow, and completions are powerful, but agents replacing developers entirely? No. Absolutely not. Today, I think that AI coding agents can absolutely replace developers. And the reason that I believe this is Claude Opus 4.5."

Burke Holland is on the train. While it's hard to separate pure hype from genuine excitement for a transformative technology, the trend is clear - the improvement and adoption curve of agentic coding tools has hit critical mass, and experienced coders are achieving quality results by integrating these tools into their workflow. Projects that used to take weeks now take days or even hours. That's exciting.

Claude Code and What Comes Next

"This is Claude Code at work, one of a new generation of AI coding tools that represent a sudden capability leap in AI in the past month or so. What makes these new tools suddenly powerful is not one breakthrough, but a combination of two advances. First, the latest AIs are capable of doing far more work autonomously while self-correcting many of their errors, especially in programming tasks. Second, the AIs are being given an “agentic harness” of tools and approaches that they can use to solve problems in new ways. The result of these two factors has led to big leaps in the latest AI tools made by the big AI companies."

Ethan Mollick gives a great overview of his experience with Claude Code, Claude Desktop and how he squeezes the most out of them.

LLM predictions for 2026, shared with Oxide and Friends

"In 2023, saying that LLMs write garbage code was entirely correct. For most of 2024 that stayed true. In 2025 that changed, but you could be forgiven for continuing to hold out. In 2026 the quality of LLM-generated code will become impossible to deny."

Simon Willison of the famous 'Pelican Riding a Bicycle' LLM test shares predictions for 2026 and beyond. As someone who writes about LLMs everyday (and as the co-creator of Django), I tend to take his thoughts seriously.

The Emperor Has No Clothes: How to Code Claude Code in 200 Lines of Code

As amazing as the frontier coding models are, it's useful to remember that the models themselves are doing the really heavy lifting - the trillion parameter models that cost hundreds of millions of dollars to train - and the 'harnesses' like Claude Code and Codex are really just small pieces of software that provide instructions, hooks and tool calls to the underlying model. Mihail Eric provides a useful example here, although at this point the harnesses are a little more complex (and making a nice TUI is a whole adventure in its own right..).

Related: Claude Code Github Repo