Twilight Techie

Neural networks whisper their secrets to me. Don't worry, I translate. #AI #DataWhisperer | Also, I may or may not be living in a simulation, because why not?

[Twilight Techie @Twilight_Techie]

This is actually pretty funny 😂

[Alex A.C. @AnalysisOp]

@SemiAnalysis_ 😅 not my vid, but it's appropriately funny

[Twilight Techie @Twilight_Techie]

@ClaudeDevs Y'all should really do a free tier, you know?

[techstash @btwiambot]

Most students say they don't have resources.

Meanwhile, some of the best learning materials on the internet are completely free:

Computer Science Fundamentals
https://t.co/TixUJ1isY3

Data Structures & Algorithms
https://t.co/EGyon2SBWu

System Design
https://t.co/3z6U7AK9Rc

Web Development
https://t.co/w0fxW0A5gW

Frontend Development
https://t.co/y9RSRdhyBt

Backend Development
https://t.co/y9RSRdhyBt

DevOps
https://t.co/y9RSRdhyBt

Cloud Computing
https://t.co/y9RSRdhyBt

Data Engineering
https://t.co/g0OljYCkD6

Machine Learning & AI
https://t.co/HiJQXjkxoC

MLOps
https://t.co/jij6IbQE4P

Cybersecurity
https://t.co/HMxTjR3Wti

Linux
https://t.co/B1uEY4w3Nl

Free Programming Books
https://t.co/syyEvqZPd5

A student with an internet connection can learn almost every in-demand tech skill for free.

Information isn't scarce.

Consistency is.

[Abhishek🌱 @Abhishekcur]

This article is literally wow.

i read it 2 years ago, and coming back to it today, it still feels new.

few tutorials teach computers in a way that permanently changes how you think. this is one of them.

If you've never built a VM before, you're missing one of the biggest "aha" moments in computer science.

[Twilight Techie @Twilight_Techie]

@quxiaoyin Alright! Will try that. Thanks 👍

[Twilight Techie @Twilight_Techie]

This is actually pretty dope! And designers think they're not getting replaced soon 😅

[Miracle Edet Okon @MimiTheDesigner]

See what chat gpt did in few seconds…a full brand identity 🙄

[Twilight Techie @Twilight_Techie]

No cap 🧢

[Sahil @sahill_og]

Claude: Just say go, and I will generate the code

Me: Go

Claude: You have reached your limit. Your limit will reset at 11:30 pm

[Twilight Techie @Twilight_Techie]

@OloriOfOloris There's not a single manual to life. Some people do better with less distractions, others with supportive people around them and frankly, if he really did grind the last couple years, just because he isn't doing better financially now doesn't mean he won't be in the long run.

[Twilight Techie @Twilight_Techie]

And that's a wrap for 2025... 2026 is gonna be fireee 🔥

[shirish @shiri_shh]

AI wrapped 2025

[Mohammed Tawakkal Ahmed @Tawakkalah13_10]

List For backend developers

• HTTP methods
• HTTP Status Codes
• Authentication OAuth, OIDC
• JWT vs Sessions
• Hashing, Salting
• Encryption standards
• Rate Limiting & DDoS protection
• WebSockets & Real-time communication
• Databases & Storage
• SQL vs NoSQL
• ACID Properties & Transactions
• Indexing strategies
• Query Optimization
• Normalization vs Denormalization
• System Design &. Architecture
• Monolith vs Microservices
• Load Balancing
• Caching Strategies
• Redis, Memcached, CDNs
• Message Brokers (Kafka, RabbitMQ)
• Docker & Containerization
• CI/CD Pipelines (GitHub Actions, GitLab CI)
• Linux Command Line & Bash scripting
• Basic Cloud Services (AWS/GCP/Azure)
• Vertical vs Horizontal Scaling
• Database Sharding & Replication
• Connection Pooling
• Unit, Integration, and Load Testing (k6, JMeter)
• Logging practices (Structured logging)
• Monitoring & Alerts (Prometheus, Grafana)
• Distributed Tracing

[Abhishek Singh @0xlelouch_]

Backend Request Lifecycle – Simplified

Client request
↓
DNS → resolves service location
↓
Load Balancer → traffic distribution
↓
API Gateway → auth, rate limits, routing
↓
Middleware → logging, tracing, validation
↓
Controller / Handler → request intent
↓
Service Layer → business logic
↓
Cache lookup → fast path (Redis / in-memory)
↓
Database query → slow path (source of truth)
↓
Transactions → consistency guarantees
↓
External calls → retries, timeouts, fallbacks
↓
Domain events → async side effects
↓
Response assembly → DTOs, serialization
↓
HTTP response → status + payload
↓
Client rendering → perceived performance
↓
Metrics → latency, error rates
↓
Logs → what happened
↓
Traces → where time was spent
↓
Alerts → when things go wrong
↓
Post-mortems → learning loop

🎯 If you understand this flow, backend systems become debuggable, scalable, and predictable.

Feel free to add what you think I missed 👇

[Nikki Siapno @NikkiSiapno]

If you want to become a better software engineer (in 2026),

read these 12 engineering blogs:

1. Meta Engineering
↳ https://t.co/Gk1odu0G1G

2. Netflix TechBlog
↳ https://t.co/T3StVaZlCB

3. AWS Architecture
↳ https://t.co/kvBAMbpyvr

4. Microsoft Engineering
↳ https://t.co/zgLHpGKBRa

5. Google Research
↳ https://t.co/UUM2DzSKQR

6. Slack Engineering
↳ https://t.co/qT1O4xxSoc

7. Discord Engineering
↳ https://t.co/ne9lNeUeMX

8. NVIDIA Developer
↳ https://t.co/i9y3zNuNyC

9. Stripe Engineering
↳ https://t.co/w8c7fcJKgj

10. Uber Engineering
↳ https://t.co/beFKMaK5r1

11. Cloudflare Blog
↳ https://t.co/oSp5ALRFPT

12. GitHub Engineering
↳ https://t.co/83WGnNkagK

What other blogs should be on this list?

--
👋 PS: Want my System Design Handbook for free?
Want my Architecture Patterns Playbook for free?
Join my newsletter with 26,501+ software engineers: https://t.co/OQtXb4zMoe

--
📌 Save for later.
♻️ Repost to help others grow.
➕ Follow Nikki Siapno + turn on notifications.

[Boris Cherny @bcherny]

I'm Boris and I created Claude Code. Lots of people have asked how I use Claude Code, so I wanted to show off my setup a bit.

My setup might be surprisingly vanilla! Claude Code works great out of the box, so I personally don't customize it much. There is no one correct way to use Claude Code: we intentionally build it in a way that you can use it, customize it, and hack it however you like. Each person on the Claude Code team uses it very differently.

So, here goes.

[Ben Awad @benawad]

Software Engineering Expectations for 2026

- The majority of your code should be written by AI now
- Cursor/Codex/Claude Code/Gemini/etc
- You should try all the tooling and switch between them, as each one gets an edge over the others depending on the release cycle.
- You should be using AI to check the code that is written by AI
- Have AI write tests
- Have AI read logs
- Have AI navigate your browser
- I don't do this every time because sometimes it's simple enough to check it myself
- You should still skim code changes
- This can be a lighter skim on internal tools and a heavier read through on customer facing code
- Use AI to help you define specs

[Akhilesh Mishra @livingdevops]

❌ DevOps is NOT:
• Kubernetes
• Jenkins  
• Terraform
• Docker
• AWS/Azure

✅ DevOps IS:
• Applications that deploy without drama
• Infrastructure that scales while you sleep
• Systems that rarely break
• Problems caught before customers notice
• Teams that actually talk to each other

Tools are just hammers and nails.
DevOps is building the damn house.

Master the WHY, not just the HOW.

Because when Kubernetes becomes obsolete (and it will eventually), your DevOps skills won't.

[Abhishek Singh @0xlelouch_]

❌ Backend is NOT:

1. A specific language or framework
2. CRUD APIs glued together
3. Writing controllers only
4. ORMs and annotations
5. Chasing the newest tech stack

✅ Backend IS:

1. Systems that stay correct under load
2. APIs that fail gracefully and predictably
3. Data models that survive product changes
4. Services that handle partial failure without panic
5. Latency, throughput, and cost kept under control
6. Clear contracts between services and teams
7. Boring, reliable systems that run at 3 a.m.

Languages and frameworks are just tools.
Backend is designing systems that hold up under load and bad external actors.

Master the fundamentals, not just the buzzwords.
So that when today’s framework is replaced and when AI takes over, your backend skills will be valid still.

[Karan🧋 @kmeanskaran]

❌ Machine Learning is NOT:

1. Coding in Jupyter Notebook
2. High accuracy, better returns
3. Model building and evaluation
4. A permanent solution for business
5. Just using tools and frameworks

✅ Machine Learning is:

1. Understanding data and feature engineering
2. A mixture of stats, maths, and probability
3. Building a system for business impact
4. An iterative approach with no single best solution
5. All about experiments, research, trial and error

Straight to the point: nowadays AI can code and provide better solutions for your ML projects. It can also write code using the latest frameworks, but in reality everything must align with the business use case.

It's already a costly side of engineering, and data is sensitive as well. You can use many algorithms for one solution, but select those that are easy to set up, have decent performance, and are compatible with your system.

There's a rule in ML:
The more fancy stuff you do, the more you get stuck in a loop of errors.

[Mihail Eric @mihail_eric]

All assignments for Stanford's The Modern Software Developer are now available online.

This is the first comprehensive university course covering how coding LLMs are transforming every stage of the software development life cycle. The assignments are intended to take you from noob to expert in how to use AI to improve your software engineering productivity.

Enjoy!

https://t.co/YxzneOSCY5

[Twilight Techie @Twilight_Techie]

@r0ck3t23 Whoa! This is a pretty big step towards efficient resource management and reducing inference times in LLMs. Big congratulations to the @Microsoft team for a job well done 👏