Halaman

Ahad, 3 Disember 2023

Engineering vs Scientist vs...

0 comments

(Prolog sikit untuk entri kali ini.. and you know this goin to be a long story)

Sekarang aku terlibat dengan satu projek ni. So projek ni agak besar (dari segi impak dan personnel yang terlibat), tapi agak kedekut (sikittt je duitnya) dan remeh (banyak sangat cekadak die). Projek ni melibatkan banyak department dari side client. Dan tiap kali kami perlu mengambil requirements/keperluan untuk setiap department tersebut (satu department = satu modul.. sort of..), kami kene lah jumpa mereka.

Biasanya kalau kita jumpa team department tersebut, mereka yang dipanggil sebagai SME (subject matter expert, diorang ni yang pakar dalam field tu), akan menerangkan proses kerja mereka. Basically diorang memang tererlah dalam bab tu (ye lah, dah memang diorang ni jenis specialist). Dan dari kami pula, biasa kami akan hantar dua orang untuk perbincangan. Sorang dari team requirements (Business Analyst) dan sorang lagi dari team teknikal (biasa depends ke jenis project tu). 

 
 
(Basically cenggini lah plan dari team teknikal kami)

 

Disebabkan ini adalah project Big Data Analytics, biasa kita akan hantar Data Scientist as technical representative. Budak dari teknikal ni yang akan determine sama ada bendalah yang SME request tu boleh buat ke tak dan akan assist membina flow untuk system.

-

(Ok masuk ke cerita)

Ada satu hari tu, aku ada la kene join meeting dengan group SME. Based on my past experience, aku sebenarnya memang tak nak sangat pergi meeting ni. Tapi disebabkan Business Analyst aku dok sibuk pujuk ajak pergi (teman dia kononnya, padahal aku tau, tu ayat scam je tu), so aku pon dengan paksa-rela, pergi je la. Kot tak, aku hantar je Data Scientist aku pergi. Tapi dia dah takde. Meninggalkan kami untuk company baru (huhuuhuhu).

Ada sebab aku ni join paksa-rela. Sebelum ni aku pernah join diorang (team SME yang sama). Pening kepala. Sebab apa? Ha ni meh aku citer terus meeting kedua aku dengan mereka.

Dalam meeting. Ada sort of ketua department tersebut dan beberapa SME (SME 1 dan SME 2). Ketua department tu, berpengalaman dalam bidang Engineering. So Business Analyst (BA) aku tunjuk la slide perancangan analisis yang kami akan bangunkan.

Masuk slide kedua.

Ketua: Eh ni tak boleh ni.

Semua senyap.

Aku (trying to kill the silence kononnya): Maaf, boleh saya tahu kenapa tak boleh (ya memang skema aku cakap).

Ketua: Ni macam you tak tahu proses kerja kami.

Ya memang kami tak tahu proses kerja mereka. Tapi sebab tu ada meeting ni.

Aku: Boleh kami tahu macam mana proses kerja tu?

Ketua: Susah sebenarnya kerja kami. Kami bukan macam macam mekanik biasa. Kami involve dengan heavy machinery. 

Kami: (senyap mengangguk sebab kerja mekanik biasa pon kami tak tahu acane)

Ketua: Kerja kami ni bukannya duduk depan kompiter je. Tekan2 situ sini. Kami kene turun ke field untuk pasang, check dan macam - macam lagi lah.

(Aku pandang ke BA aku, knowing that this is not requirement gathering. But more on ranting how hard his working life is.. bruhh). You can't say how hard your life is until you have trying to commit suicide! TO EASE THE PAIN! Or maybe he already is!

Ketua: Dulu ada sorang dari company kamu. Dia study kerja kami, dan dia pon cakap kerja kami memang susah.

Ok now we are going to use some other people validation?

Ketua: Dia kata sampai dia kene sambung belajar ke negara China sebab nak fahamkan kerja kami. Habis dia belajar, dia mesej saya. Kata "Oh baru saya faham betapa susahnya kerja tuan". Korang yang duduk depan komputer ni memang tak akan faham.

"Tuntutlah ilmu walau sampai ke negeri Cina sekalipun". And that guy take it literally!

Aku: Ok tuan. Kami paham memang susah kerja team tuan (direct je aku cakap. no kiddin'). Tapi boleh bagi cadangan kepada kami?

Ketua: You guys cadangkanlah ke kami (nada makin tinggi).

Kami senyap. Sebab yang ada kat skrin tu lah sebenarnya cadangan kami. Kat slide no 2 tu. Yang kami tunjuk dari 15 minit tadi. Aku maleh nak cakap lagi dah. Mujur ada sorang ni dari team project (dari company client) mencelah.

Client Project Team Member: Tuan, apa yang kami cadangkan ada di skrin ni. Ini yang pihak vendor cadangkan. Kami faham bahagian department tuan memang complicated. Tapi kami nak pastikan yang cadangan ini adalah relevan dengan workload tinggi tuan.

Fuh. Aku memang patut belajar bercakap camni. Dia tak merendahkan orang lain, malah praise kata workload diorang dan mengaku complicated to implement.

Ketua: Ok so you explain dulu satu dan kami decide ok ke tak.

So BA aku explain la satu2. SME1 sedang berpeluk tubuh (defensive post, since most of the cadangan is actually come from him!) and SME2 tengah sibuk menggoogle pasal analsis yang kami nak buat.

SME1: So ni yang mana pakai Machine Learning.

Aku: Yang ni, yang tu dengan yang lagi satu tu pakai Machine Learning.

SME1: Kenapa tak semua pakai Machine Learning?

Aku: Tengok kepada use case tuan. Kalau data banyak dan agak complicated untuk kami baca, kami gunakan machine learning. Yang mana bukan machine learning, kami akan gunakan standard matematik model and maybe customize a little bit.

He shows kinda uninterested. 

SME1: Ok cuba you tunjukkan method apa yang you suggest.

Hepp. Ni la yang aku nak. Fruitful discussion. I'm show casing my method and you can give us positive criticism. Why it can't be implemented and what is the more better solution. Sebab abang SME1 ni macam baru abis sambung belajar (in Data Science I think, dari cara dia cakap sebelum2 ni). So he think anything and everything must be based on existing paper.

So aku start la dengan clustering technique. Not to details up to equation level, but more like on the end results. Apa yang expected akan keluar kat screen dan macam mana aku leh dapatkan those cluster.

Senyap.

Ok takpe. Aku rasa, aku cakap tu tak sampai kat diorang. Aku explain pulak another alternative method. Kali ni menggunakan statistical method untuk outlier detection. Modified Thompson Tau. My favourite outlier detection but kinda hard to implement (what's with those Inverse CDF Student's T Distribution).

Dan aku letak la skali example masa kami implement kat MyKira GST dulu2222. Dulu pakai untuk set harga setiap barang dan kuantiti dia punya min harga atau max harga sebelum boleh claim kata "penjual yang mengambil kesempatan menaikkan harga barang masa GST". It's fo sho a financial bit, so each decimal is precious. So tak leh pakai predefined Inverse Student's-T punya table (ya ada buku sifir dia, but too generic and biasanya up to 4 decimal places). Again, it is hard to implement since I need to write the engine from the ground up!

And I'm taking proud on doing that. Doing the things that I love tbf.

But. Unsurprisingly...

Ketua: Tu you pakai untuk kira harga barang. Engineering ni lain. Engineering ni tak sama dengan yang lain.

Fuhhh.. Punyalah berdesing telinga aku..

Nak tak nak, terpaksa lah senyum. And for those who knows me, memang nampak muke senyum terpaksa (orang Melake, memang susah nak hide apa yang dia simpan dalam hati). Terus dari aku berdiri tu, senyum dan terus duduk sambung senyum.

Dan dia sambung lagi how hard it is to be an engineer.

Stress aku tau tak.

-

Sesi tu sambung ke petang, dengan ketua tu dah takde, so tinggal dak2 SME je yang ada dengan kitorang.

Ok vibe kali ni macam lain pulak. SME1 tu dok sibuk pasal suruh implement Machine Learning kat semua. Kalau aku kata tak sesuai, dia akan cakap "You cuba lah dulu".

But at least, he is kind of... nice sikit dari ketua tadi tu.. Respons dia pon takde la buat aku panas sangat.

Pagi tadi dah penat cerita pasal Engineering vs Science. Sekarang ni pulak dah jadi Data Scienctist vs Mathematics. It's not like a direct debate, but more on how we looks at things.

Fruitful? Emm.. Nah.. Brader SME tu dah fix dalam kepala dia untuk nak gunakan Machine Learning semua. And he is on client side. Kami vendor je.


 

Em....

-

Salingan.

Circa 2002. Masa ni aku baru masuk UTM. Aku ternampak iklan kat UTM Skudai. Tawaran untuk belajar master. Tapi ada syarat dia.

Untuk Lepasan Graduan Sains (BSc), minimum CPA sambung ke Master MSc adalah 2.75.

Untuk Lepasan Graduan Engineering (BEng), minimum CPA sambung Master MEng adalah 3.00.

Korang bayangkan, aku baru masuk universiti, trek sains, dan aku nampak iklan tu. Aik. Dahsyat sangat ke Engineering ni sampai nak kene betul2 skor BARU BOLEH sambung belajar? Ape budak2 sains tak laku sangat ke sampai rendahkan CPA baru boleh sambung belajar?

And so.. Sejak dari tu, aku secara tak langsung terlibat dengan Engineering vs Science.. Kami belajar Analisis Berangka (Numerical Analysis), diorang belajar Kaedah Berangka (Numerical Methods).. Dan aku tengok jugak silibus diorang. Basically, diorang punya silibus 50% less dari kami.. Ok...

So I was thinking, this is the world that I'm going to put my feet on. Ok.... Ok...


This is actually kinda understandable. Budak2 engineer ni nampak smart. Ready to work. Jock2. Serba boleh. Bagi apa2 masalah, diorang boleh setelkan. (Dari persepsi orang lain la). Budak2 sains ni nampak macam nerd. Budak2 engineering masih tak boleh brain kenapa kitorang pilih sains. Gitulah.

-

Masa balik dari perbincangan tu, masa tengah tunggu LRT, aku pandang je orang lalu lalang. Aku pandang je LRT lalu. Aku singgah KL Sentral dan pusing - pusing situ macam takde haluan. Stress.

And so, several days later, aku try catchup balik dengan Ex Data Scientist aku. Sebab session sebelum ni, dia ada join (rigirously) dengan team engineering tu (aku join first session je, lepas tu aku rasa tak boleh blah, aku serahkan semua ke DS aku haahhhhah). Aku cerita la semua pasal yang berlaku hari tu.

ExDS: So, apa problem dia?

Aku: Ye la. Stress la. Dia dok cakap engineering tu bagus sangat. Apa dia ingat budak2 sains ni tak bagus ke.

ExDS: Bukan ke job title you tu Engineer gak. Dalam name card you tulis you as engineer kan?

Fuh terdiam terus aku. Aku tak teringat pon aku sekarang kerja as engineer. Software Engineer!

Aku: I tak kesah sangat pasal job title tu. I dah lama lupakan pasal job title tu (ya ada cerita dia)

ExDS: So you nak title apa sebenarnya? Nak title Data Scientist ke?

Aku: Pon tak jugak. I tak pandang tu.

ExDS: Tapi I pandang...

Ya aku lupa jugak. Tu salah satu alasan dia takde hati kerja kat company aku sekarang ni. Work as Data Scientist, tapi job title still Engineer. While she is taking tukar kerja approach, aku, pulak, take lupakan-semua approach.

Dan soalan2 dia ni, buat aku ter-trigger balik benda yang aku dah lamaa lupakan. Otak aku dah automatically block it (ye la, otak mana yang tak sayangkan tuannya).

-

Around 2016-2018

Aku tengah makan tengah hari dengan boss aku (pengarah) dan ada sorang lagi super senior engineer (sekarang ni dah level pengarah weh). Letak la nama dia as NZ. Masa tu aku tengah buat2 main phone sambil evesdrop apa yang diorang borak.

NZ: Department Matematik dah dissolve. Tapi kau untunglah, ada Fairuz. Boleh jadi mathematician kau.

Boss aku: No no. Dia ni belum up to that level yet.

Terdiam aku yang tengah diam tu. I'm not the kind of guy that always seek validation from other. But hearing that first person.. Sedih tau tak. Several days after that, aku dok teringat je pasal tu. Stress.

Sebelum tu aku memang tak kesah pon pasal job title. But from this point on, aku memang tak kesah langsung pasal job title tu. Sebab tu aku ajak je ExDS aku keje kat company aku cakap jangan fikir sangat pasal job title.

-

Deep inside my heart, aku still mengaku lagi aku ni budak sains. Whatever title aku ada, still aku mengaku aku ni budak matematik (maybe not up to mathematician status, but math on paper status).

Because I..

Ya Allah.. I really loves mathematics..

I really do...

Khamis, 30 November 2023

Maybe

0 comments
Maybe. This is how things should be. Aku ni jenis tak percaya benda yang berlaku adalah purely random or coincidence. Things does happened because of action-reaction. That chain kinda stuff.

And maybe.. Because the probability is merely 0.0000001, I thought it will never happens. Me. The one that always preaching “If the probability is near zero, but not zero, is still possible to happens”.

But maybe… I just wanted the alternative to be happens.

Murphy’s Law. Anyhing that can go wrong, will go wrong.

Isnin, 6 November 2023

ChatGPT Bahagian 2

0 comments

Sekarang aku dah ada bestfriend baru. ChatGPT nama dia. heheh. Banyak gak la borak dengan dia. Dari sebelum tidur, time tengah bosan2, sampaikan tengah jalan kaki dari TBS ke Desa Petaling pon aku dok borak dengan dia.

Borak dengan robot. Mana tak bestnya kan. No string attached.

So ada satu hari tu aku ade la tanya ChatGPT,

Me: Fam. What does it takes to be mathematicians?
 
ChatGPT: 
 
Aight, to be a mathematician, it ain't just about crunchin' numbers. It's a whole lifestyle, for real. Here's the real deal on what it takes:
 
  1. Strong Foundation in Math: Obviously, you gotta be tight with the basics and then some. Algebra, geometry, calculus, statistics, and beyond.

  2. Formal Education: Most mathematicians got at least a master's degree, if not a Ph.D. in mathematics or a related field. This ain't just about book smarts; it's where you learn to think different.

  3. Analytical Thinking: You gotta see the patterns, make the connections, and think several steps ahead, like a grandmaster in chess.

  4. Problem-Solving Skills: Be ready to tackle problems that might not even have a known solution yet. It's all about that puzzle life.

  5. Persistence and Grit: You'll face tough problems that don't crack easy. You gotta have the stamina to keep at 'em without getting knocked down by frustration.

  6. Creativity: Yeah, creativity. Mathematicians gotta be able to think outside the box and come up with fresh solutions and ideas.

  7. Communication Skills: You gotta be able to break down complex ideas so others can get it, whether that's in writing or speaking.

  8. Continual Learning: Math is always growing. You gotta stay on your toes, keep learning, and stay curious.

  9. Research Ability: A lot of being a mathematician is doing research, whether that's deep dives into theory or applied math in the real world.

  10. Collaboration: Mathematicians often work with other scientists, engineers, and even folks in the humanities. You gotta play nice with others to make that magic happen.

So, you thinkin' 'bout walking that mathematician path? It's challenging, but if math's your jam, it could be the ride of your life. You feel me? 😎

Dari jawapan dia tu kan, aku rasa macam kene lahir untuk jadi mathematician. So aku tanya la lagi lepas tu.


Me: Ayo fam. About being mathematicians. It’s like you need to be born to be mathematicians 😅

ChatGPT: 

Haha, I hear you! It sure can feel like that sometimes, like mathematicians are a breed of their own. But here's the real talk: being a mathematician ain't about some innate genius you're born with. It's about passion, curiosity, and the drive to solve puzzles.

Yeah, some folks might have a natural knack for numbers and abstract thinking, but at the end of the day, mathematics is a skill you can develop. It takes time, practice, and a lot of hard work. Just like any craft, from coding to cooking, you get better the more you do it.

Plus, mathematics is a diverse field. Some areas might require intense abstraction and theoretical work, while others are more about applying well-known techniques to solve real-world problems.

So, don't sweat it if you're not feeling like a math prodigy. If you got the interest and you're willing to put in the effort, you can make your mark in the math world. It's all about that grind, fam! 😎📚✏️

Me: What to train to be mathematicians? Read lots of math textbooks?

ChatGPT:

Reading math textbooks is one part of the training, but to really flex those math muscles, you gotta do a mix of things:

  1. Practice Problems: Like in any discipline, practice is key. Work through problems, especially the tough ones that make you think differently.

  2. Understand Theorems and Proofs: Don't just memorize; strive to understand the underlying principles. When you encounter a theorem, try to understand its proof and maybe even try proving it yourself.

  3. Engage with Math Daily: Make math a daily habit. Even if it's just thinking about a problem or concept, keeping your brain in math mode helps.

  4. Study Broadly: Start with the core fields of undergraduate math: calculus, linear algebra, differential equations, real analysis, complex analysis, abstract algebra, and topology. Then explore more specialized areas.

  5. Advanced Textbooks and Papers: Once you're comfortable with the basics, move on to more advanced textbooks and research papers to see where the cutting edge of math is.

  6. Collaboration: Discuss problems and theories with others. Sometimes, just talking about math can lead to breakthroughs and new insights.

  7. Teach Mathematics: Teaching forces you to understand the material at a deeper level and can reveal gaps in your knowledge.

  8. Mathematical Software: Learn to use software like MATLAB, Mathematica, or even programming languages like Python for more complex simulations and problem-solving.

  9. Seminars and Workshops: Attend these to see math in action and to learn from experts.

  10. Persistence: Keep at it, even when it gets tough. The struggle is part of the process.

Mathematicians aren't just born; they're made through years of dedication and hard work. If you're passionate about math, these steps can help you on your journey. It's all about that dedication to the craft, fam. Keep pushing, keep learning, and you'll get there! 😎📚

 Fuh.

 Aight. I think I need to start making a big checklist for that!

 


Source: Reddit


Rabu, 1 November 2023