Code
Bhavya
Build one focused systems project, test it against known results and explain its design under interview pressure. A small correct project with measurements is stronger than a large unexplained clone.
Select a card to generate its evidence plan.
Compare FCFS, SJF/SRTF, Priority and Round Robin using a shared workload and measurable fairness/latency metrics.
Build command execution, redirection and pipelines around the Linux process/file-descriptor APIs.
Translate virtual addresses and compare FIFO, LRU and Optimal page replacement with an auditable trace.
You will receive the minimum evidence required for a credible portfolio demonstration.
Define valid input, output, constraints and failure behavior. Include at least three acceptance examples with known answers.
Keep algorithms separate from DOM or terminal code. Pure functions make edge-case testing easier.
Show intermediate states—ready queue, descriptors, frames or page table—so a reviewer can audit the decision.
Cover empty values, duplicates, limits, invalid data, ties and algorithm-specific edge cases.
Report a meaningful metric, compare configurations and explain what the number does not prove.
Document architecture, assumptions, test commands, screenshots, limitations and the next improvement.
| Artifact | What it proves | Minimum standard |
|---|---|---|
| README | Communication and reproducibility | Problem, architecture, setup, examples, tests, limits |
| Test suite | Correctness discipline | Known-answer cases plus invalid/edge inputs |
| Trace or visualizer | Understanding of internal state | Every transition has a cause and labeled result |
| Measurements | Ability to evaluate a design | Workload, metric, result and interpretation |
| Commit history | Incremental ownership | Small meaningful commits, no secrets or generated noise |
| Demo script | Interview readiness | Two-minute happy path plus one failure case |
claim + test + observed result = evidence
trade-off = benefit + cost + context
limitation stated honestly builds trust
Program/process, PCB, states, context switch, thread models, scheduler metrics, starvation and aging.
Race condition, critical-section requirements, mutex/semaphore/monitor, deadlock conditions and recovery.
Paging, segmentation, TLB, page faults, replacement, locality, working set and thrashing.
Inodes, allocation, free space, journaling, interrupts, DMA, disk scheduling and RAID trade-offs.
Authentication/authorization, reference monitor, ACL/capability, least privilege and defence in depth.
fork/exec/wait, pipes, signals, descriptors, VM/container choice, namespaces and cgroups.
Mark complete after your project has a README, tests, trace and two-minute demo.
Saved in this browser.