Every few years the industry rediscovers that programming languages are not religions. Then we immediately behave like they are religions. Someone posts a benchmark. Someone else says memory safety. Someone says developer experience. A distributed systems person appears from under a bridge and whispers “Erlang solved this in 1998.” A startup founder announces they are rewriting their CRUD app in R
If this is useful, a ❤️ helps others find it. I run both in production. Here's the real comparison — not theoretical, from actual use building developer tools. Local LLM (Ollama) Gemini API (Free) Cost $0 forever $0 (free tier) Privacy 100% local Data sent to Google Setup Install Ollama + pull model Get API key (2 min) Quality Good (7B), Great (70B) Excellent Speed Fast if model lo
Becoming a tech lead was the goal from pretty early in my career. I had a clear picture of what the role was. More responsibility, more influence over the work, more of the interesting problems landing on my desk because someone had to figure them out and that someone, finally, would be me. It read like the natural next step. The thing you graduate to once you're good enough. What that picture did
A College Project That Planted a Seed Years ago I was on a university team trying to build a Go AI. We explored monte carlo simulation for lookahead search, basic neural networks for pattern recognition, and expert systems for encoding domain knowledge. None of them worked well enough on their own. Go's branching factor is enormous, so brute-force search fails quickly. Neural networks without th
DeepClaude: I Combined Claude Code with DeepSeek V4 Pro in My Agent Loop and the Numbers Threw Me Off DeepSeek V4 Pro correctly solves 94% of deep reasoning tasks in my loop… but the latency cost makes it unusable for 60% of my agent cases. Yeah, you read that right. And that completely blows up the narrative of "combining models is always better." Tuesday night I watched the DeepClaude post cli
Series: AI Isn’t an Engineering Problem Anymore (Part 2) In the last post, I talked about hitting a usage limit while debugging my robot and realizing how repetitive my own AI usage had become. When we use LLMs, whether through APIs or tools, it feels like every request is new. The inefficiency isn’t from using AI too much. You don’t ask once, you iterate. These are the most interesting ones. Some
Reaching an annual salary of ¥8,000,000 is often seen as a major milestone for software engineers in Japan in 2026. On paper, it sounds like a ticket to a comfortable, upper-middle-class life in Tokyo. But is 8 million yen a good salary in Tokyo—really? But if you are coming from abroad—or if you've only looked at the "Gross" figure on your offer letter—you might be walking into a "logic bug" that
Run the same brand-query through ChatGPT, Gemini, Perplexity, Claude, and Grok. Read the citations. The cited URLs will not be the same, the brands featured will not be the same, and in roughly a third of cases one tool will cite your brand confidently while another does not mention it at all. The temptation is to reach for an algorithmic explanation different rerankers, different summarisation st