now
Writing more than shipping. Focusing on clarity in long-form notes.
digital diary
hey, i am sanjay
Building systems by day, collecting quiet ideas by night. This page is my personal operating log, not a polished resume.
now page
learning
Data contracts, model evaluation design, and better technical storytelling.
letting go
Overcomplicated dashboards, fragmented task systems, and noisy workflows.
philosophy
தெய்வத்தான் ஆகா தெனினும் முயற்சிதன் மெய்வருத்தக் கூலி தரும்.
Even if fate doesn't favor you, effort will reward the toil of your body.
Thirukural 619
my journey in tech
freshman
Computer Science Engineering
internship
Software Engineer
graduate
Computer Science Engineering
full-time
Software Engineer
machine learning engineer
Building and shipping ML systems
reading shelf
my work gear
This is how my setup look now
device
for the nerds
-
primary machine
ASUS TUF Gaming A15, Ryzen 5 4600H (6C/12T), 15.6-inch FHD up to 144Hz, 16GB DDR4 RAM, Nvidia GeForce GTX 1650, NVMe SSD setup.
-
monitor
LG UltraGear 27-inch gaming monitor, IPS QHD 2560x1440, 144Hz, 1ms mode, HDR10, G-SYNC Compatible, FreeSync Premium.
-
keyboard
Cosmic Byte CB-GK-40 Artemis, 68-key, tri-mode (wired + 2.4GHz + Bluetooth), per-key RGB, hot-swappable switches.
-
primary mouse
Amkette Hush Pro Epic M, triple-mode wireless, DPI steps 800/1200/1600/3200, rechargeable 400mAh class battery.
-
secondary mouse
MFTEK wired USB mouse, LED backlit, up to 2000 DPI.
-
pen tablet
Huion HS64, 6.3 x 4 inch active area, 8192 pressure levels, 5080 LPI.
-
microphone
MAONO PD200X dynamic microphone with boom arm, dual USB-C and XLR output, cardioid pickup, up to 24-bit/48kHz via USB.
-
earpods
OnePlus Nord Buds 3, 12.4mm driver, up to 43 hours total playback.
software stack
confession
reality
-
my code lives in
VS Code (because everyone's extensions are here now) -
primary language
Python (with occasional tourist visits to Go and C++)
-
I promise myself I'll learn
Vim keybindings (currently on day 847 of 'tomorrow')
-
tabs vs spaces
Tabs. I said what I said. (except YAML, obviously)
-
my most-opened app
-
what actually trains my models
PyTorch + copious amounts of coffee + prayer
-
container situation
Docker (47GB of cached nightmares)
-
cloud playground
AWS for everything (SageMaker handles ML workflows, IAM for character building) -
regex patterns
-
database choice
PostgreSQL (until I need vectors, then Milvus)
-
note-taking system
Obsidian (300 files, 12 are useful)
-
kubernetes experience
We're working through some YAML trauma
-
git commits
'fix' → 'actually fix' → 'wtf' → 'please work'
recently played
live from last.fm
say hi
If this resonates, feel free to reach out.
Email: heissanjay@proton.me