At 2:17 AM, my monitoring alert yanked me out of sleep: the customer service bot had suddenly lost its memory. Users were asking “Where is my order?” three times in a row, and it kept asking for their phone number as if they were complete strangers. I opened the logs and saw that ConversationBufferMemory was loading empty message lists. The key was still there in Redis, but somehow deserialization
I finished an English series on the way I think ordinary people can start using AI for real work. The point is not to become an AI expert first. The point is to have one place where you can say what you want, give the tool access to the right folder, and check the result. Anything important still needs a human pause: publishing, deleting, paying, or authorizing. My preferred starting point is simp
If this is useful, a ❤️ helps others find it. All tests run on an 8-year-old MacBook Air. HiyokoLogcat supports Japanese and English. The AI diagnosis needed to respond in whichever language the user chose. The simplest solution: write the system prompt in the target language. Gemini follows it reliably. // Don't do this let prompt = format!("Analyze this log: {}\nRespond in Japanese.", context);
37 days. That's how long the main and submain branches diverged before the big merge today. It wasn't just about closing this gap; it was about making the biggest forward leap we've seen in weeks. The test matrix exploded from 78 to 117 tests, and we dropped an 11-commit sprint into IR lowering that hammered out essential struct and array support. That alone makes you want to take a closer look at
When building modern applications, one problem shows up everywhere: How do I uniquely identify data across systems? That’s where UUIDs (Universally Unique Identifiers) come in. A UUID is a 128-bit unique identifier used to identify information in distributed systems. Example: 550e8400-e29b-41d4-a716-446655440000 It looks random - and that’s the point. Traditional IDs (like auto-increment integers
If this is useful, a ❤️ helps others find it. All tests run on an 8-year-old MacBook Air. If a user closes the AI diagnosis overlay and reopens it, should you call Gemini again? No. Cache the result. Same input → same output. No reason to burn rate limit quota. Here's the caching layer I built into HiyokoLogcat. Without caching: User clicks diagnose on error line 847 Gemini responds in 3 seconds U
It happened to me. My site — bashsnippets.xyz — had been down for six hours before I knew That's the kind of thing that sits with you. The fix took 20 minutes to build. Here's exactly what I wrote and why The natural instinct is to just open a browser and load the page. You only check when you remember to You don't check at 2am when it actually goes down What you want is automated, logged, and
"Nobody tells freshers the real numbers. HR says 'competitive salary.' LinkedIn shows fake CTCs. College placement cells lie. This post doesn't." Let me ask you something uncomfortable. 👇 You're about to graduate. You're applying for jobs. You see a role that says "CTC: 6-12 LPA." What does that actually mean for your bank account every month? 🤔 What's the difference between a ₹6 LPA offer in Ba