Software Engineer | SRM'26
Code your future๐ฉโ๐ป
Have a positive mindset ๐ง
Keep learning ๐
Spread your wings & fly ๐ชฝ
Life is meant to be beautiful โจ
The judging lines are full. The projects are ready.
We asked hackers on the floor at ETHGlobal Mumbai one simple question:
What did you build in 36 hours?
The ReAct framework combines reasoning and acting, forcing an AI agent to alternate between thinking out loud (Chain-of-Thought) and executing external actions (like web searches or code execution).
This iterative loop is "Thought-Action-Observation" loop.
#AI#Agents
Your agent is not autonomous
You are its loop
You check every step because nothing else in the system can
I broke down the 7 levels from babysitting agents to running a fleet โ https://t.co/4RacdXA4nE
AI Agents: Beginner to Advanced Guide
Download here: https://t.co/Jm3LZN2PHT
A practical guide to AI agents, automation, tools, and building intelligent systems.
Want to build your skills in data and AI? Start here:
Step 1: Azure Data Fundamentals (DP-900)
Step 2: SQL AI Developer Associate (DP-800)
Step 3: Fabric Data Engineer Associate (DP-700)
Resources every software engineer should know about ๐
Not courses.
Not certifications.
Just stuff worth bookmarking.
1. Cloudflare Learning
DNS, CDN, networking, security, HTTP
https://t.co/cxCvFwA5W5
2. Google Engineering Practices
How engineers actually write and review code
https://t.co/fKMgnqaTMd
3. Uber Engineering
Real systems, scaling, infra + AI
https://t.co/u899psbwaj
4. AWS Architecture Center
Real architecture patterns + diagrams
https://t.co/IuYXKE6Zy4
5. Stripe Engineering
Payments, APIs, distributed systems
https://t.co/P6YdFM1IIT
6. Netflix TechBlog
Distributed systems, data + infrastructure
https://t.co/Tf6cGaMLiL
7. GitHub Engineering
How GitHub builds and scales its platform
https://t.co/KGcI35mWD2
8. Meta Engineering
Production ML + infra at massive scale
https://t.co/4MeoU8rsSB
Bookmark these. ๐
Reading how companies solve real engineering problems is a completely different kind of learning.
Are you aware of these full forms?
๐ซ GPT: Generative Pre-Trained Transfomer
๐ซ BERT: Bidirection Encoder Representations From Transformer
#ai#gpt#techgyaan
๐ AI Prompt Types
๐น Zero-shot: Direct instruction, no examples.
๐น Few-shot: Learn from examples.
๐น Chain of Thought: Reason step by step.
๐น Zero-shot CoT: Direct prompt + reasoning.
๐ก Better prompts = better AI responses.
#AI#PromptEngineering#LLMs