LOG-009
◇ IN THIS LOG
◇ SERIES PATH
◉ 0 READS
● OPEN TO CLOUD/DEVOPS ROLES/◉ NEW YORK, NY
Arafat Sani
AI Infra, In Public
IT SUPPORT SPECIALIST → AI INFRASTRUCTURE
— IT support specialist climbing into AI infrastructure. I run models on affordable GPUs, learn Kubernetes, Terraform and CUDA in the open, and log every step.
NOW →Working toward the CKA while running open models on affordable cloud GPUs.
AI INFRASTRUCTURECLOUDINFERENCEMLOPSDEVOPSPLATFORMML INFRANETWORK
◉ LEARNING LOGS
⊗ BUILDS DOCUMENTED
⁂ YEARS IN OPS
00
Honest logs from the climb into AI infra
LOG.ALL
00
Small builds I actually ran and wrote up
STK.DEPLOYED
00+
Years across help desk, desktop & NOC-style ops
OPS.YEARS
◈ FEATURED LOGLOG-009 / INFERENCE
Serving LLMs at 4x:
vLLM, KV-Cache
& Autoscaling
Inference engineering field log — how continuous batching + PagedAttention took a 7B model from 18 → 74 tok/s on one A10G, and how I autoscale it on Kubernetes without burning budget.
009
Learning Logs
Published
Published
TRACK 01 / RUNBOOKS ⊕
07
Builds
Documented
Documented
TRACK 02 / BUILDS ✕
◉ INFRA KNOWLEDGE BASE — 01
Learning Logs
& Field Notes
Filter by engineering track. Add your own tags, delete ones you don't use — the filter bar is yours. Tags persist locally.
Click a pill to filter. Click × on a pill to delete it.
52RUNBOOKS
LOG.ALL
LOG.ALL
09TRACKS ACTIVE
TRK.INFRA
TRK.INFRA
04CERTS EARNED
NVIDIA + NEBIUS
NVIDIA + NEBIUS
07:12NEXT WINDOW
LIVE.REC
LIVE.REC
▣ BUILD EVIDENCE — 02
Infrastructure
Builds
Small, real and reproducible. Every build is something I ran, broke, measured or wrote up — notes included, hype excluded.
◉ 07BUILDS DOCUMENTED
✳ 09LEARNING LOGS
◉ DOCUMENTED
IN PUBLIC.
IN PUBLIC.
▣ BLD — BUILDS, RUNS & EXPERIMENTS
LOGGED // REPRODUCIBLE
BUILDS →
Serving, scheduling,
cloud & delivery.
Small builds, real runs.
◉ WORK HISTORY — 03
Experience
& Roles
Roles, teams, and what shipped. Full dossier with skills, certs and education lives in About — this is the short path.
◉ TOOLBOX — 04
Tools I
Actually Use
No affiliate links, no sponsored picks. Every tool here runs something of mine — with a one-line verdict on why it earned its place.
This list grows as the builds do — check back.
Infra stack
◈ I SANI.LOG — 05
Learning
Signal Index
0 Words
Logged in public
Logged in public
◉ MY ROADMAP INTO AI INFRASTRUCTURE
Road Into
AI Infrastructure
◈ | SANI.LOG
Done: Support Roots & Degrees
Now → Next: GPUs, K8s & Inference
◌4+ Years IT Support & OpsDONE — REAL
◉M.S. Data Science, GPA 3.9DONE — 2025
◉Nvidia + Nebius GPU CertsDONE — 2026
⚠Kubernetes & Terraform BasicsLEARNING — NOW
◎CUDA & Inference DepthNEXT UP
⧉AI Infra / Inference RoleTHE GOAL
ID // OPERATOR FILE
Arafat Sani
AI Infrastructure / Inference Engineer
FROM
Tier 1 Triage→TO
Reliable Ops
Tier 1 Triage→TO
Reliable Ops
IT Support Specialist with B.S. Computer Science, M.S. Data Science and 4+ years across help desk, desktop, service desk and NOC-style operations.
09
STACK: WINDOWS 11 / MACOS / MICROSOFT 365 / SERVICENOW / ACTIVE DIRECTORY / TCP/IP / DNS
4+ Years Across Help Desk & NOC Ops ◉
Support Coverage Focus
Windows 10/11, macOS/iOS, Microsoft 365 and Google Workspace, Active Directory/MFA access, LAN/Wi-Fi/VPN and TCP/IP/DNS/DHCP, ServiceNow ticketing and knowledge-base documentation.
PLATFORM AUTHORITY
WINDOWS 10/11, MACOS & IOS90%
MICROSOFT 365 & GOOGLE WORKSPACE88%
TICKETING & SLA OWNERSHIP87%
- ◉ Service Desk & Ticketing Unit
- ◈ Systems / Network Unit
CONTINUE →
AS
IMG // OPERATOR
◉ FULL OPERATOR DOSSIER — EDITABLE IN ADMIN
Arafat Sani
AI Infrastructure / Inference Engineer
AI infrastructure & inference engineer with an M.S. in Data Science and 4+ years of ops roots across help desk and NOC-style environments.
▣ SKILLS // PROFICIENCY
▣ TECH STACK
▣ CERTIFICATIONS — EARNED + NEXT
▣ EDUCATION
INFRA COLLABORATION
Let's
Work Together
◈ Hiring for cloud support, junior DevOps/MLOps or GPU-adjacent roles — or just want to talk homelabs and inference? I reply within 48 hours.
◇ DISCUSSION