Shreyash

Spent this weekend coding fully by hand for the first time in almost a year. No agents. No AI autocomplete. Just docs open in Chrome, insomnia for firing requests, and Zed. Honestly, the slow part isn’t writing business logic. It’s all the scaffolding and maintenance work around it.

What changed is this: If you deeply understand the architecture and can structure the problem clearly, agents become insanely effective. At that point, coding is no longer the bottleneck. Thinking is.

Spent this weekend coding fully by hand for the first time in almost a year. No agents. No AI autocomplete. Just docs open in Chrome, insomnia for firing requests, and Zed. Honestly, the slow part isn’t writing business logic. It’s all the scaffolding and maintenance work around it.

1/ Ever watched your servers start sweating because a single rogue bot or a buggy client retry loop went wild and hammered your API with hundreds of requests per second? That’s rate limiting in action – the invisible guardrail that keeps your system alive, fair, and costs in check.

11/ Tips • Dynamic limits per tier ? Pull from Consul/etcd. • Log every 429 → spot abuse patterns early. • Test race conditions and window boundaries like your on-call depends on it (it does). • Real war story: badly tuned rate limiters have taken down major services during flash sales.

1/ Ever watched your servers start sweating because a single rogue bot or a buggy client retry loop went wild and hammered your API with hundreds of requests per second? That’s rate limiting in action – the invisible guardrail that keeps your system alive, fair, and costs in check.

9/ Placement is the real engineering decision: • Edge/CDN → kill junk before it hits your VPC • API Gateway → cleanest for most teams • Server-side → only when you need deep per-user business logic Layer them. That’s how you think like someone who actually ships.

1/ Ever watched your servers start sweating because a single rogue bot or a buggy client retry loop went wild and hammered your API with hundreds of requests per second? That’s rate limiting in action – the invisible guardrail that keeps your system alive, fair, and costs in check.

Sliding Window Counter is accurate enough for production, low memory, no timestamp spam.

1/ Ever watched your servers start sweating because a single rogue bot or a buggy client retry loop went wild and hammered your API with hundreds of requests per second? That’s rate limiting in action – the invisible guardrail that keeps your system alive, fair, and costs in check.

6/ 4. Sliding Window Log Store a timestamp for every single req. On new req, drop anything older than the window & count the rest. Superpower: Most accurate – no boundary tricks possible. Catch: Memory explodes at scale (millions of timestamps = pain) Only when absolute fairness is non-negotiable.

1/ Ever watched your servers start sweating because a single rogue bot or a buggy client retry loop went wild and hammered your API with hundreds of requests per second? That’s rate limiting in action – the invisible guardrail that keeps your system alive, fair, and costs in check.

4/ 2. Leaky Bucket Requests pour in but only “leak out” to your servers at a fixed constant rate. Bucket fills up? Extra requests get dropped. Superpower: Perfectly smooth downstream traffic – your DB never sees spikes. Catch: Zero bursts allowed. Use when stability > flexibility.

1/ Ever watched your servers start sweating because a single rogue bot or a buggy client retry loop went wild and hammered your API with hundreds of requests per second? That’s rate limiting in action – the invisible guardrail that keeps your system alive, fair, and costs in check.

2/ There are exactly 5 algorithms you must know. I’ll give you the mental model, when to pick each, and the real trade-offs (the stuff interviewers actually test).

This sounds amazing! 🔥 I’ve seen many projects using a combo of tools like esbuild and the TypeScript compiler, so with tsc speeding up, I’m sure we’ll see simpler, faster builds across the board. Thanks @typescriptlang.org team

After years of selectively reading in tech, Machine Learning with Pytorch and Scikit Learn is my first book read cover to cover! (except the math, I needed a second round for that 😅) Thankyou @sebastianraschka.com, I guess this is the most well-rounded and comprehensive book on Machine Learning

After 5 years of on-and-off learning, I finally published my first package on crates.io It might not be super useful, but I’m proud of it I’m comfortable with Rust now, though work keeps me busy with JS, Python & Go 😅 Would love to build some devtools and OSS