ryanthegiantlion

🧠 How to Use LLMs Effectively for Coding (2025 Edition)

LLM-assisted coding has evolved dramatically. If you're coming from the Copilot autocomplete era, the modern agentic workflow will feel like a superpower. This guide outlines practical patterns that work in real-world software engineering.

---

1. Create Project Context Before Doing Anything Else

Agentic LLMs only work well when they understand your project.

ryanthegiantlion

Various variables are made available by postman runtime and can be references in the test script. Common ones are responseCode and responseBody and responseTime

• Parse response body var data = JSON.parse(responseBody);

• Test Format tests["Test Name"] = {condition} e.g tests["Has token"] = data.data.login != null;

• Response times tests["Response time is less than 500ms"] = responseTime < 500;

ryanthegiantlion

Resetting pg_stat_database and pg_stat_all_tables

To reset below states

select pg_stat_reset();

SELECT pg_stat_statements_reset()

ryanthegiantlion

{
  "waves": [
    {
      "_id": "5805bc78a64d73001b8fcdc6",
      "user": "5805b6a5a64d73001b8fcda6",
      "content": "Let's make some waves!",
      "track": "5805bc78a64d73001b8fcdc4",
      "date": "2016-10-18T06:08:35.000Z",
      "__v": 0,
      "geo": {

ryanthegiantlion

show dbs // list all indexes
use <db>
show collections
db.collection.find(<query>) // find given optional query
db.directmessages.remove({}) // delete all
> use mydb; 
> db.dropDatabase();
db.foo.update({}, {$set: {lastLookedAt: Date.now() / 1000}}, { multi: true } // update all 

ryanthegiantlion

mongo:

mongod

redis:

redis-server

rabbit:

rabbitmq-server

ES:

ryanthegiantlion

curl -v -H "Accept: application/json" -H "Content-type: application/json" -X POST -d '{"userName": "ryan", "platform":"android", "pushToken":"egO6G6CoxZE:APA91bFqN90Lb0n4TfNC02dXJ0vMT7uQ33WqFBG_FU5AC1dW_5PX1t9yFdA6gBxG_THDHdkkFdSZXq05PCABiIfAHUhu9ky-4vq8MOJdMUIZ1p4RvwiTVcAqnBGi3je19cO7TporEgjx"}'  http://localhost:8082/users/

ryanthegiantlion

// setup

1. if you want the worker to be in a separate project and separately deployed: heroku create <projectname>
2. if separate project created in step 1: heroku addons:attach <addonname>

// subscriber

var redis = require('redis');
var RedisChannels = require('./constants/redisChannels')

ryanthegiantlion

• Disk failures: http://static.googleusercontent.com/media/research.google.com/en//archive/disk_failures.pdf
• Databases architecture: http://db.cs.berkeley.edu/papers/fntdb07-architecture.pdf
• Dynamo DB http://www.allthingsdistributed.com/files/amazon-dynamo-sosp2007.pdf
• Google File System http://static.googleusercontent.com/media/research.google.com/en//archive/gfs-sosp2003.pdf
• Pub sub architecture http://people.mpi-inf.mpg.de/~vsetty/Setty_Phd_Thesis.pdf
• Pub usb spotify https://www.csc.kth.se/~gkreitz/spotifypubsub/spotifypubsub.pdf

ryanthegiantlion

import sys

T = int(sys.stdin.readline().strip())

xor_bits = (1 << 32) - 1

for i in xrange(T):
    N = int(sys.stdin.readline().strip())
    print N ^ xor_bits