Assembly, C, and Rust
Instructions, registers, memory, ownership, performance, and the hardware–software boundary.
00 systems / assembly / rust / linux / infrastructure
Systems first. Platforms built to endure.
I’m Ishant Kumar Das. This is my working systems portfolio, technical cheat sheet, and learning surface across hardware, low-level software, Linux, virtualization, backend platforms, cloud engineering, accelerated compute, and AI infrastructure.
01 flagship direction
From instruction set to cloud platform.
The center of gravity is systems engineering: understanding each layer well enough to operate it, debug it, and build reliable layers above it.
Linux, VMs, and Docker
Operating systems, isolation, networking, containers, self-hosting, and operational recovery.
Backend, cloud, and AI
Reliable services and compute pipelines with portable boundaries across Cloudflare, AWS, and GCP.
02 technical cluster map
Related systems, kept together.
Technologies are grouped by the system they help build. A language can appear in more than one cluster when it serves a different engineering role.
Systems core
The primary discipline: instructions, memory, kernels, low-level software, and the behavior of machines under real constraints.
- Assembly
- C
- Rust
- Linux
- Operating systems
- Computer architecture
Platform and virtualization
The operating layer that turns machines into dependable platforms: VMs, containers, networking, observability, and self-hosted infrastructure.
- Virtual machines
- Docker
- Containers
- Linux
- Networking
- Reliability
Accelerated computing
GPU-aware computation from data layout to kernels, keeping numerical correctness and execution evidence connected.
- CUDA
- C++
- Rust
- Matrices
- Tensors
- GPU scheduling
Backend and cloud engineering
APIs, data, automation, and portable cloud adapters. Cloudflare is home; AWS, GCP, and self-hosted paths remain deliberate options.
- Backend systems
- SQL
- JSON
- Cloudflare
- AWS
- GCP
- Git
- CI/CD
AI infrastructure
Model serving, evaluation, monitoring, lifecycle governance, and accelerator-adjacent infrastructure treated as systems problems.
- AI systems
- ML pipelines
- LLM infrastructure
- Observability
- Governance
Web delivery
A thin, accessible presentation layer for exposing systems work. It supports the portfolio; it is not the portfolio's technical center.
- HTML
- CSS
- JavaScript
- Astro
- Accessibility
03 selected evidence
Systems work, recorded.
Self-Hosted Systems Lab
A Linux-centered environment for understanding virtual machines, containers, networking, service isolation, observability, and controlled failure.
- Linux
- Virtual machines
- Docker
- Containers
- Networking
- Backend engineering
- Git
- CI/CD
Tensor Systems Lab
A cross-language laboratory for matrix and tensor execution across C++, Rust, C, and CUDA kernels.
- CUDA
- C++
- Rust
- C
- Matrices
- Tensors
- JSON
04 learning architecture
Six ordered tracks.
Each track moves from concepts to experiments, then into artifacts that can be inspected and tested.
- 01
Systems core
The primary discipline: instructions, memory, kernels, low-level software, and the behavior of machines under real constraints.
flagship- Assembly
- C
- Rust
- Linux
- Operating systems
- Computer architecture
- 02
Platform and virtualization
The operating layer that turns machines into dependable platforms: VMs, containers, networking, observability, and self-hosted infrastructure.
flagship- Virtual machines
- Docker
- Containers
- Linux
- Networking
- Reliability
- 03
Accelerated computing
GPU-aware computation from data layout to kernels, keeping numerical correctness and execution evidence connected.
specialization- CUDA
- C++
- Rust
- Matrices
- Tensors
- GPU scheduling
- 04
Backend and cloud engineering
APIs, data, automation, and portable cloud adapters. Cloudflare is home; AWS, GCP, and self-hosted paths remain deliberate options.
platform extension- Backend systems
- SQL
- JSON
- Cloudflare
- AWS
- GCP
- Git
- CI/CD
- 05
AI infrastructure
Model serving, evaluation, monitoring, lifecycle governance, and accelerator-adjacent infrastructure treated as systems problems.
systems application- AI systems
- ML pipelines
- LLM infrastructure
- Observability
- Governance
- 06
Web delivery
A thin, accessible presentation layer for exposing systems work. It supports the portfolio; it is not the portfolio's technical center.
supporting interface- HTML
- CSS
- JavaScript
- Astro
- Accessibility
05 knowledge records
A cheat sheet that grows into shared insight.
Systems core
From Instruction Set to Cloud Platform
A systems map connecting machine instructions, memory, kernels, Linux, virtual machines, containers, backend services, and cloud infrastructure.
Accelerated computing
Matrix Memory Layout
Working notes on row-major storage, logical shapes, physical strides, contiguous views, and the cost of traversal order.
Short references support fast recall. Connected explanations preserve the reasoning behind them, so personal learning can become useful guidance for others.
Public summaries are separate from future protected notes, source artifacts, and personal learning records.
06 evidence model
Experience organized by system, not by buzzword.
The full stack page expands each cluster into concrete evidence states: exploring, practiced, built, shipped, or maintained.
Inspect the clustered evidence matrix