Muel (Dio)

junior data , dukun & design.
and i wonder if you know what it means

[Muel (Dio) @designthesign]

luck is what happens when preparation meets opportunity

[GenZGrind @GenZ_grind]

Gue ada experience, buat fresh graduate.

Bayangin lo baru lulus kuliah 11 bulan lalu. CV lo masih tipis, pengalaman kerja cuma magang 3 bulan, tapi lo udah nerima slip gaji **Rp 13,5 juta** per bulan + bonus tahunan 2 bulan + full WFH + allowance HP & internet.

Itu bukan mimpi. Itu temen gue. Biar gua kasih tipsnya, nanti.

[Elysia🪷 @ElysiaXBT]

Lo baru sadar HPnya ilang, langsung buka Find My Device lewat laptop.

Last seen 5 jam lalu di Stasiun Gambir. Abis itu? Udah.

Nggak gerak lagi. Langsung offline.

Si maling langsung matiin internet biar ga ke trace. Itu cara jadul.

Dulu sampe situ doang.
Sekarang? Nggak lagi.

[Kr @karirfess]

Kalian tau berita ini kan???

Korupsi Silmy Karim di Ditjen Imigrasi: Rp145,5 Miliar✅

Kasus korupsi di Direktorat Jenderal Imigrasi semakin menghebohkan setelah KPK menetapkan Silmy Karim sebagai tersangka.

- Total Kerugian Negara: Rp145,5 miliar😋
- Periode Korupsi: Berlangsung sejak 2022 hingga 2026 (kurang lebih 4 tahun)😩

Modusnya tuh gini:
- Pemerasan terhadap Warga Negara Asing (terutama WNA China) yang mengurus KITAS dan KITAP.
- Memperlambat proses izin tinggal, lalu meminta "uang pelicin" atau "biaya percepatan".
- Uang hasil pemerasan dibagikan secara rutin setiap hari Jumat ke para oknum di pusat dan daerah.

Bayangin aja, orang hari jumat bukannya?
- Bagi2 Jumat Berkah❌
- Jumat malah bagi hasil korupsi✅

[Elysia🪷 @ElysiaXBT]

Rumusnya namanya STAR. Bukan motivasi atau quote. Ini KERANGKA yang bikin cerita lo masuk ke kepala HRD.

S — Situation: Konteks cerita lo apa?
T — Task: Lo dikasih tanggung jawab apa?
A — Action: Lo lakuin apa secara spesifik?
R — Result: Hasilnya berapa? Ada angkanya?

Tanpa STAR = curhatan. Pake STAR = presentasi diri yang menjual.

[Muel (Dio) @designthesign]

@iyaiyanyenyenye @sweetbubblepink @worksfess role apa nder kalau boleh tau?

[Rui Ma @ruima]

https://t.co/z02N0Uq3yv

[Pradithya Aria Pura @pradithya_aria]

Gw mau cerita soal salah satu project paling wild yang pernah gue kerjain: Feast, open-source feature store yang lahir dari kebutuhan nyata di Gojek. 🧵

[Darshil | Data Engineer👨🏻‍🔧 @parmardarshil07]

I Reviewed Hundreds of PySpark Pipelines. Most Make the Same Mistakes ⬇️

Most people use PySpark.

Very few write it like professionals.

If your Spark code works but feels slow, messy, or scary to maintain, this is usually why.

Here’s what pros do differently 👇

1️⃣ They think in data volumes, not rows
If your mental model is still Python loops, you’re already in trouble.

Pros design transformations assuming millions or billions of records.

Rule of thumb:
If it can be expressed as a transformation, it should be.

2️⃣ They control the schema early
Letting Spark infer schema on production data is gambling.

Pros:
- Define schemas explicitly
- Validate columns upfront
- Fail fast if something is off

Cleaner pipelines. Fewer 2 am surprises.

3️⃣ They avoid UDFs unless absolutely necessary
UDFs look convenient. They’re performance killers.

Pros prefer:
- Built-in Spark SQL functions
- Window functions
- Higher-order functions

If Spark can optimise it, let Spark handle it.

4️⃣ They write readable transformations, not clever ones
One giant chained expression might look smart.

It’s a nightmare to debug.

Pros:
- Break logic into logical steps
- Use meaningful column names
- Treat transformations like business logic, not puzzles

Readable code scales better than “smart” code.

5️⃣ They cache with intention, not habit
Caching everything is as bad as caching nothing.

Pros cache only when:
- The DataFrame is reused
- The computation is expensive
- Memory usage is justified

Every cache has a cost. Know why it exists.

6️⃣ They design for failure
Bad data will come. Late data will come. Duplicate data will come.

Pros write PySpark assuming:
- Nulls will exist
- Types will break
- Joins will explode rows if unchecked

Defensive coding is professional coding.

What are the other tips you'll give to write better Spark code? Let me know 👇🏻

[pdawg @prathamgrv]

Introducing Tensortonic research

> Implement ML papers in cloud-native IDEs
> Breakdown of all papers to architecture, math, and code
> State-of-the-art papers like Transformers, BERT, ViT, DDPM, VAE, GANs and many more

[Vaishnavi @_vmlops]

Y Combinator is honestly one of the best platforms to find startup jobs, internships & WFH roles across domains
I’ve seen a lot of hiring for AI/ML roles here lately
Definitely worth checking out

[Jobs with Aramide @AramideOyekunle]

Career Mapping Exercise:

– Go to LinkedIn
– Search for the job you want (e.g. “Data Analyst” or “Project Manager”)
– Look at 5 people with that title
– Study their path: education, first job, certifications
– Reverse engineer your roadmap
– Make a 12-month Notion plan

You don’t need luck. You need a blueprint.

[DataVidhya @thedatavidhya]

sql
↓
python
↓
data structures (for data, not leetcode)
↓
linux & bash
↓
git & github
↓
relational databases (postgres / mysql)
↓
data modeling (oltp vs olap)
↓
etl / elt concepts
↓
apache spark
↓
batch processing
↓
streaming (kafka)
↓
data warehouses (snowflake / bigquery / redshift)
↓
orchestration (airflow)
↓
cloud (aws / gcp / azure)
↓
data lakes (s3 / gcs + parquet)
↓
data quality & monitoring
↓
basic devops (docker)
↓
ci/cd for data pipelines
congrats, you’re a data engineer now.

[Muel (Dio) @designthesign]

@adxtyahq how about using UI kit?

[khairunnisa pramana @nyssasya]

amal jariyah detected

[neural nets. @cneuralnetwork]

Hey

Go try out - https://t.co/jctAuMJtJw

[Krishna Agrawal @Krishnasagrawal]

AI agents staircase 📘📚

[Abraham John 🦄🦓 @Abmankendrick]

UI/UX Designers, you can now create unlimited animated gradient videos without paying a single cent.

I just found this website called Krumzi, a simple but powerful animated background generator that lets you create smooth, eye-catching gradient videos for free. Perfect for landing pages and hero sections.

Can’t believe it’s free; bookmark it for later. 💜

[Kevin Sebastian @kevinsebast_]

Gue sekarang mau praktik langsung yaa tentang cheat interview pake STAR METHOD:

S – Situation
Context singkat aja, ceritain lo kerja dimana, jabatan lo apa, di kerjaan lama lo, fokusnya diminta apa?
“Misal saya jadi head of sales di company abc, posisinya waktu itu, Tim sales lagi drop conversion.”

T – Task
Peran lo apa di situ.
Bukan tugas tim, tapi tanggung jawab lo personal misal personal KPI nya apa"

“Gue diminta improve conversion di tahap discovery.”

[Muel (Dio) @designthesign]

@toolfolio cool, what recording tools do u use btw?