Interview practice plan

C

Software Engineer | Distributed System + Fullstack (Python,Go, Java) + AI - 7+ years

Cisco · Bengaluru

Data & AIMid level31 practice questions

Practice content written by LYDJ for interview preparation. These are common questions for this kind of role, not this employer's actual interview questions.

What this role asks for

Skills we found in the job description. The plan below is built around them.

PythonGoJavaGenerative AIDistributed systemsProduct managementSRETypeScriptJavaScriptRustAgileLLMsMicroservicesKafkaRedisDocker

Also named in the description, not covered below: SRE, Agile, Kafka, Redis, Docker, Kubernetes, Observability, CI/CD. Review these on your own.

Sign in to see your gap areas first. We compare this role with the skills on your profile.

0 of 31 questions practised

Answer each one out loud, then tick it. Progress is saved in this browser only.

01Required skill0/3

Python

This role asks for Python.

  • What are generators, and when would you use one instead of building a list?

    What good looks like

    Lazy evaluation with yield, constant memory for large or infinite streams, and pipelines of generators; note they can only be iterated once.

  • What goes wrong with def add(item, items=[])?

    What good looks like

    Default arguments are evaluated once, so the list is shared across calls. Use None as the default and create the list inside.

  • How do you structure a Python codebase so a team can work on it safely?

    What good looks like

    Locked dependencies, pyproject-based packaging, type hints checked in CI, pytest with fixtures, linting and formatting, and clear module boundaries.

02Required skill0/3

Java

This role asks for Java.

  • What is the difference between checked and unchecked exceptions, and how do you decide which to throw?

    What good looks like

    Checked for recoverable conditions the caller must handle, unchecked for programming errors. Mention wrapping, not swallowing exceptions, and keeping the cause.

  • Explain the Java Memory Model. What do volatile and synchronized actually guarantee?

    What good looks like

    Visibility versus atomicity, happens-before, and why a volatile counter++ is still not thread-safe (use atomics or locks).

  • Your Java service throws OutOfMemoryError in production about once a week. How do you investigate?

    What good looks like

    Heap dump on OOM, analyse retained sizes and dominators, check caches and unbounded collections, and correlate with traffic or batch jobs.

03Required skill0/3

LLMs & GenAI

This role asks for Generative AI and LLMs. These LLMs & GenAI questions cover the ground they share.

  • Explain retrieval-augmented generation. When does it beat fine-tuning?

    What good looks like

    RAG grounds answers in fresh, citable documents without retraining; fine-tuning changes style or format behaviour. Many systems use both.

  • How would you evaluate an LLM-powered feature before launch?

    What good looks like

    A golden dataset of real tasks, rubric-based and automated checks, human review of samples, a regression suite for prompt changes, and live feedback.

  • How do you reduce hallucinations in a question-answering system over company documents?

    What good looks like

    Improve retrieval (chunking, hybrid search, reranking), instruct the model to answer only from sources with citations, allow it to abstain, and measure with evals.

04Required skill0/3

Product management

This role asks for Product management.

  • A key metric dropped right after your launch. What do you do?

    What good looks like

    Verify the data, check whether the launch caused it (cohorts, rollout groups), decide whether to roll back, and communicate a plan with timelines.

  • How do you define success metrics for a new feature?

    What good looks like

    Link to the product's north-star metric, choose a leading indicator you can move, add guardrail metrics, and make sure it can be measured and attributed.

  • Write the one-page brief for a feature that lets candidates track their job applications. What goes in it?

    What good looks like

    Problem and evidence, target users, goals and non-goals, success metrics, key user flows, risks and open questions.

05Required skill0/3

TypeScript

This role asks for TypeScript.

  • Explain generics with constraints. Write a typed function that picks a set of keys from an object.

    What good looks like

    Use <T, K extends keyof T> returning Pick<T, K>, and explain how the constraint keeps callers from passing keys that do not exist.

  • What are discriminated unions, and how do they help with exhaustive checks?

    What good looks like

    A shared literal field narrows the type in a switch; assigning the default case to never makes the compiler flag any unhandled variant.

  • When is using any or a type assertion acceptable, and how do you keep a large codebase strictly typed?

    What good looks like

    Prefer unknown plus narrowing, enable strict mode, validate external data at the boundary, and tighten gradually with lint rules.

06Required skill0/3

JavaScript

This role asks for JavaScript.

  • Explain the event loop. In what order do setTimeout, a resolved Promise and code after await run?

    What good looks like

    Call stack, microtask queue (promises) drained before the next macrotask (timers), and how await splits a function into a microtask continuation.

  • What is a closure? Give a practical use and a bug closures can cause.

    What good looks like

    A function remembering its lexical scope; uses include private state and factories; classic bug is var in a loop or a stale value captured in a callback.

  • Implement debounce, and explain when you would use debounce versus throttle.

    What good looks like

    Debounce waits for a pause (search input), throttle limits the rate (scroll handlers). Handle this, arguments and clearing the timer.

07Role function0/3

Machine Learning

This is a data science / ML role. Expect questions on modelling choices, evaluation and production ML.

  • Your classifier has 98% accuracy but the business says it is useless. What might be going on?

    What good looks like

    Class imbalance. Look at precision, recall, PR-AUC and the confusion matrix, the cost of each error type, and threshold tuning.

  • How would you deploy a model, and how would you know it still works three months later?

    What good looks like

    Versioned models and features, shadow or canary rollout, monitoring input and prediction drift, tracking performance as labels arrive, and retraining triggers.

  • How do gradient-boosted trees work, and when would you prefer them over a neural network?

    What good looks like

    Sequential trees fitting residuals; strong on tabular data with modest size, fast to train and easier to explain. Neural nets win on images, text and huge datasets.

08Fundamentals0/3

Core programming

This role asks for Go and Rust. We do not have questions specific to these yet, so practise these fundamentals and answer them in that language.

  • Explain the difference between a process and a thread, and when you would use each.

    What good looks like

    Memory isolation versus shared memory, cost of creation and context switches, and the synchronisation problems that sharing introduces.

  • Given an array of integers and a target, return the indices of two numbers that add up to the target. Then improve the complexity.

    What good looks like

    Start with the O(n^2) brute force, move to a single pass with a hash map for O(n), and talk through duplicates and the no-answer case before coding.

  • Walk me through what happens when you type a URL into a browser and press Enter.

    What good looks like

    DNS, TCP and TLS handshakes, the HTTP request, server processing, then parsing, rendering and loading sub-resources. Depth in one area beats a shallow list.

09Design round0/3

System design

This role asks for Distributed systems and Microservices, so expect an architecture round. State requirements and numbers first, then trade-offs.

  • Design a rate limiter for a public API.

    What good looks like

    Compare token bucket, fixed and sliding windows; decide where it lives (gateway or service), how counters are shared across nodes, and how clients learn they are limited.

  • Design an upload and processing pipeline for large documents such as resumes.

    What good looks like

    Direct-to-storage uploads with signed URLs, async processing through a queue, retries and a dead-letter queue, malware scanning, and status updates for the user.

  • Design the REST API for a multi-tenant SaaS product.

    What good looks like

    Resource modelling, versioning, pagination, authentication and tenant isolation, idempotent writes, and a consistent error contract.

10Every loop0/4

Behavioural

Every interview loop has a behavioural round. At mid level, expect questions on delivering independently, handling ambiguity and working through disagreement. Use the STAR structure.

  • Describe a piece of feedback that changed how you work.

    What good looks like

    State the feedback, your honest first reaction and the concrete behaviour you changed. Close with evidence that the change stuck.

  • Tell me about a time a project you owned was at risk of missing its date.

    What good looks like

    Show early detection, honest escalation and the scope, sequencing or resourcing levers you pulled. Avoid stories that end with "we all worked weekends".

  • Tell me about a time you worked closely with a team in another time zone or culture.

    What good looks like

    Mention concrete practices (async updates, overlap hours, written decisions) and a moment where a misunderstanding was caught and fixed.

  • Tell me about a time you used data to change someone's mind.

    What good looks like

    Explain what the data showed, how you made it understandable for that audience, and which decision changed as a result.

Company research brief: Cisco

From the roles we track on LYDJ, refreshed daily from the official careers page.

120

open roles

90

added in 30 days

10

in Data & AI

Product & SoftwareGCCGreat Place to Work

Hiring most for

  • Engineering71
  • Other11
  • Data & AI10
  • QA & Testing10

Cities

Bengaluru · 116Hyderabad · 2Chennai · 1Pune · 1
See all 120 roles at Cisco →

Read before the interview

Questions to ask them

  • ›There are several open Data & AI roles at Cisco right now. How is the team growing, and where does this role fit?
  • ›The company hires in Bengaluru, Hyderabad, Chennai. How do teams across locations work together day to day?
  • ›How does the team use Python and Go today, and what would you like to change?
  • ›What does success look like in this role after six months?
  • ›What is the biggest challenge the team is facing this year?

Preparation checklist

0/9 done

Saved in this browser only.

Interview prep: Software Engineer | Distributed System + Fullstack (Python,Go, Java) + AI - 7+ years at Cisco | LYDJ