cswork / libraryOpen practice workspace

Interview guides · Rad Ai

Rad Ai Software Engineer Interview Guide

Rad Ai Software Engineer interview guide: reported rounds, questions and how candidates describe the process.

The 60-second brief

At Rad Ai, a Software Engineer plays a pivotal role in transforming the radiology workflow by building highly reliable, scalable, and intelligent systems. The engineering team is responsible for developing products that directly reduce radiologist burnout and improve patient care outcomes. As a engineer here, you will work at the intersection of generative AI, complex clinical workflows, and high-performance cloud infrastructure, making your contributions highly visible and impactful. The systems you design and maintain must process massive amounts of medical data with extreme precision and minimal latency. You will collaborate closely with machine learning researchers, product managers, and clinical experts to translate advanced AI models into seamless, intuitive applications.

This guide is scoped to a Software Engineer candidate at Rad Ai.

Focus areas

  • TypeScript
  • Python
  • Systems design

The Rad Ai interview process

Rad Ai candidates report 4 rounds over 3-5 weeks. The stages below are what candidates describe, not a published process.

Recruiter Screen

Initial discussion to align on background, career goals, and expectations.

What to demonstrate

  • Initial discussion to align on background, career goals, and expectations
  • Depth in TypeScript

How to prepare

  • Be able to walk your CV end to end in two minutes, and say why this company specifically.
  • Have your salary expectations, notice period and location constraints ready, and ask for the rest of the loop in writing.

Take-Home Coding Challenge

Candidates complete a comprehensive coding challenge within a 48-hour window.

What to demonstrate

  • Candidates complete a comprehensive coding challenge within a 48-hour window
  • Depth in TypeScript

How to prepare

  • Answer aloud and timed: Write a clean, modular service in Python to interface with an external AI model API, incorporating retry logic and error-handling boundaries.
  • Answer aloud and timed: Refactor a highly coupled legacy function to improve its testability, readability, and performance.

Code Review Session

Detailed review of the take-home coding challenge with the team.

What to demonstrate

  • Detailed review of the take-home coding challenge with the team
  • Depth in TypeScript

How to prepare

  • Answer aloud and timed: Design an asynchronous task-processing queue capable of handling high-throughput medical image analysis requests.
  • Answer aloud and timed: How would you structure a real-time collaborative editing platform for clinical reports, ensuring data consistency and low latency?

Systems Design Interview

Interview with a principal architect or engineering leader focusing on system design.

What to demonstrate

  • Interview with a principal architect or engineering leader focusing on system design
  • Depth in TypeScript

How to prepare

  • Answer aloud and timed: Design a scalable API gateway that handles authentication, rate limiting, and request routing for multiple microservices.
  • Answer aloud and timed: How would you architect a system to cache and deliver large-scale machine learning model outputs efficiently to a global client base?

Where candidates lose points

Prioritize Quality Over Completion

If you cannot finish every single feature of the take-home within a realistic timeframe, focus on making the core functionality exceptionally robust, clean, and well-tested. Document any uncompleted features in your README.

Review System Design Fundamentals

Refresh your knowledge of caching strategies, load balancing, database scaling, and asynchronous message queues before your system design round.

Be Ready for Interactive Coding

During the code review, you may be asked to modify your take-home code on the fly. Ensure your local environment is set up and ready to run the project smoothly.

Do not copy-paste boilerplate code or use unverified third-party libraries extensively in your take-home

The engineers will drill deep into your code during the review, and you must be able to explain every single line you submitted.

Questions & practice

Implement a robust data-parsing pipeline in TypeScript that processes asynchronous inputs and handles partial

Medium · Coding & Practical Implementation

Implement a robust data-parsing pipeline in TypeScript that processes asynchronous inputs and handles partial failures gracefully.

Approach

  1. Say what the runtime actually does before reasoning about the code.
  2. Name what is shared across threads and what owns each piece of state.
  3. Identify the window where an invariant is briefly untrue.
  4. Distinguish a value from a reference to it, and say which one you handed out.

Follow-up

  • What happens if two callers reach this at the same time?
  • Where could this allocate more than you expect?

Create a reusable full-stack component that fetches, filters, and displays real-time data while optimizing for

Medium · Coding & Practical Implementation

Create a reusable full-stack component that fetches, filters, and displays real-time data while optimizing for minimal network overhead.

Approach

  1. Restate the input: its shape, its size, and what is guaranteed about it.
  2. Name the brute-force solution and its complexity before improving on it.
  3. Choose the data structure from the access pattern, not from familiarity.
  4. State the target complexity and say which constraint rules the naive version out.

Follow-up

  • How does this change if the input no longer fits in memory?
  • What is the worst case, and how likely is it on real data?

Refactor a highly coupled legacy function to improve its testability, readability, and performance.

Medium · Coding & Practical Implementation

Approach

  1. Restate the input: its shape, its size, and what is guaranteed about it.
  2. Name the brute-force solution and its complexity before improving on it.
  3. Choose the data structure from the access pattern, not from familiarity.
  4. State the target complexity and say which constraint rules the naive version out.

Follow-up

  • How does this change if the input no longer fits in memory?
  • What is the worst case, and how likely is it on real data?

Write the update path that detects a concurrent edit

Medium · optimistic concurrency · transactional outbox · audit history

resource carries version INT NOT NULL DEFAULT 1. resource_revision holds revision_id, resource_id, version, actor_user_id, change_kind, patch JSONB, request_id, created_at with UNIQUE (resource_id, version). outbox_event holds aggregate_type, aggregate_id, aggregate_version, event_type, payload, status. A PUT carries the version the client read. Write the exact statements for the single transaction that applies the edit, records the revision and enqueues 'resource.updated', and give the handler's branch on zero affected rows. Then say what PostgreSQL 16 does under READ COMMITTED when two of these updates hit one row at once.

Approach

  1. One transaction, three writes, no network call inside it: UPDATE resource SET title = $3, version = version + 1, updated_at = now() WHERE resource_id = $1 AND tenant_id = $4 AND version = $2; then INSERT the resource_revision row at version $2 + 1; then INSERT the outbox_event row at the same aggregate_version. The event goes to a table rather than a broker because no transaction spans both.
  2. Branch on the affected-row count before doing anything else. Zero has three causes — stale version, wrong tenant, row gone — so re-read once and map to 409 carrying the current version, or 404 for an id outside the caller's tenant, which also stops the endpoint confirming that another tenant's id exists.
  3. State the engine behaviour instead of assuming it. Under READ COMMITTED the second UPDATE blocks on the row lock, and when the first commits PostgreSQL re-evaluates the WHERE clause against the newly committed row, so the version predicate now fails and the statement reports zero rows. Under REPEATABLE READ the identical collision raises SQLSTATE 40001 instead, so the handler must fold both shapes into one conflict response.
  4. Keep UNIQUE (resource_id, version) even though the predicate already serialises writers. It is what makes a lost update unwritable if any other path ever reaches the revision table, and it converts a logic bug into 23505 rather than into a silently missing history row.

Follow-up

  • A client sends the version it read ten minutes ago and the resource has moved three versions. What is in your 409 so it can resolve the conflict without a full re-fetch?
  • Two editors, two disjoint fields, no overlap. Does your answer still refuse the second write, and should it?

Stop tag and share joins from fanning out a page

Medium · join fan-out · semi-joins · many-to-many

resource_tag is (resource_id, tag_id) with PK (resource_id, tag_id); resource_share is (resource_id, shared_with_user_id, permission). The tagged-and-shared listing inner-joins resource to both, filters tenant_id, tag_id = ANY($2) and shared_with_user_id = $3, orders by updated_at DESC and takes 50. Pages come back with fewer than 50 distinct resources and the total in the header is far too high. Explain the row multiplication, rewrite both the page query and the count query so each is correct, and name the index each one needs. PostgreSQL 16.

Approach

  1. Do the arithmetic against the predicates that are actually there. An inner join emits one row per matching child row, and both joins are filtered: tag_id = ANY($2) admits only the requested tags, shared_with_user_id = $3 admits one user's share rows. So a resource holding three of the requested tags and shared with $3 once yields three rows, not one — the multiplier is its count of matching tags times its share rows for that single user, and that second factor is 1 unless the table admits duplicate (resource_id, shared_with_user_id) pairs. LIMIT 50 then limits rows rather than resources, and COUNT(*) counts pairs — the header is the product, not the population.
  2. Reject DISTINCT as the fix. It deduplicates after the product has been built, so the planner must materialise and sort the fanned-out set before the LIMIT can apply, and it leaves any SUM or AVG in the same select list wrong.
  3. Rewrite both filters as semi-joins, keeping resource as the only row source: AND EXISTS (SELECT 1 FROM resource_tag rt WHERE rt.resource_id = r.resource_id AND rt.tag_id = ANY($2)) and the same shape against resource_share. A semi-join stops at the first match per resource and preserves the driving index order, so ORDER BY updated_at DESC, resource_id DESC LIMIT 50 still stops after 50 rows.
  4. Count with the same predicates and no join at all: SELECT count(*) FROM resource r WHERE r.tenant_id = $1 AND r.status = 'active' AND EXISTS (...) AND EXISTS (...). Nothing multiplies a resource, so the number is the population.

Follow-up

  • The filter changes from 'any of these tags' to 'all of these tags'. Rewrite it and state what it costs relative to the ANY form.
  • A resource can be shared with the same user twice under different permissions. Does your count change, and should it?

Write a clean, modular service in Python to interface with an external AI model API, incorporating retry logic

Medium · Coding & Practical Implementation

Write a clean, modular service in Python to interface with an external AI model API, incorporating retry logic and error-handling boundaries.

Approach

  1. Say who the caller is and what they do when the call fails halfway.
  2. Define the identity of a request so a retry cannot double-apply it.
  3. Separate accepted, pending, failed and confirmed; they are different facts.
  4. Design the error taxonomy before the success shape; callers branch on it.

Follow-up

  • What happens if the caller retries after a timeout?
  • How does a client discover it is on an old version of this contract?

Design an asynchronous task-processing queue capable of handling high-throughput medical image analysis reques

Medium · System Design & Architecture

Design an asynchronous task-processing queue capable of handling high-throughput medical image analysis requests.

Approach

  1. Fix the scope first: who calls this, how often, and what they do when it fails.
  2. Name the read and write paths separately; they rarely have the same bottleneck.
  3. Choose a partition key and say what query it makes expensive.
  4. State the consistency you need, and where you are willing to be stale.

Follow-up

  • What breaks first when traffic grows ten times?
  • How does this behave when that dependency is down for an hour?

How would you structure a real-time collaborative editing platform for clinical reports, ensuring data consist

Medium · System Design & Architecture

How would you structure a real-time collaborative editing platform for clinical reports, ensuring data consistency and low latency?

Approach

  1. Fix the scope first: who calls this, how often, and what they do when it fails.
  2. Name the read and write paths separately; they rarely have the same bottleneck.
  3. Choose a partition key and say what query it makes expensive.
  4. State the consistency you need, and where you are willing to be stale.

Follow-up

  • What breaks first when traffic grows ten times?
  • How does this behave when that dependency is down for an hour?

Design a scalable API gateway that handles authentication, rate limiting, and request routing for multiple mic

Medium · System Design & Architecture

Design a scalable API gateway that handles authentication, rate limiting, and request routing for multiple microservices.

Approach

  1. Fix the scope first: who calls this, how often, and what they do when it fails.
  2. Name the read and write paths separately; they rarely have the same bottleneck.
  3. Choose a partition key and say what query it makes expensive.
  4. State the consistency you need, and where you are willing to be stale.

Follow-up

  • What breaks first when traffic grows ten times?
  • How does this behave when that dependency is down for an hour?

How would you architect a system to cache and deliver large-scale machine learning model outputs efficiently t

Medium · System Design & Architecture

How would you architect a system to cache and deliver large-scale machine learning model outputs efficiently to a global client base?

Approach

  1. Fix the scope first: who calls this, how often, and what they do when it fails.
  2. Name the read and write paths separately; they rarely have the same bottleneck.
  3. Choose a partition key and say what query it makes expensive.
  4. State the consistency you need, and where you are willing to be stale.

Follow-up

  • What breaks first when traffic grows ten times?
  • How does this behave when that dependency is down for an hour?

How do you approach learning a complex, highly specialized domain (such as medical technology or AI integratio

Medium · Behavioral & Collaboration

How do you approach learning a complex, highly specialized domain (such as medical technology or AI integration) when starting a new project?

Approach

  1. Say who the caller is and what they do when the call fails halfway.
  2. Define the identity of a request so a retry cannot double-apply it.
  3. Separate accepted, pending, failed and confirmed; they are different facts.
  4. Design the error taxonomy before the success shape; callers branch on it.

Follow-up

  • What happens if the caller retries after a timeout?
  • How does a client discover it is on an old version of this contract?

Share an experience where you diagnosed and resolved a critical production issue under intense time pressure.

Medium · Behavioral & Collaboration

Approach

  1. Establish what changed and when, before forming any theory.
  2. Pick a bisection that eliminates candidates whichever way it turns out.
  3. Check the instrumentation before believing the symptom.
  4. Separate the trigger from the cause; the deploy is rarely the bug.

Follow-up

  • What would you look at first, and what would it rule out?
  • How would you tell a cause from a coincidence here?

Suggested practice checklist

Built from the rounds and topics Rad Ai candidates report.

Day 1 — Map the Rad Ai loop

  • Write out the reported sequence: Recruiter Screen, Take-Home Coding Challenge, Code Review Session, Systems Design Interview.
  • For each round, write one sentence on what it is judging, from the description above, and mark the one you are least ready for.

Deliverable: A one-page map of the 4 reported rounds, with the weakest marked.

Day 2 — Work TypeScript

  • Spend the session on TypeScript, which Rad Ai candidates report being tested on.
  • Write one worked example in TypeScript and time yourself on it.

Deliverable: One timed worked example in TypeScript.

Day 3 — Work Python

  • Spend the session on Python, which Rad Ai candidates report being tested on.
  • Write one worked example in Python and time yourself on it.

Deliverable: One timed worked example in Python.

Day 4 — Work Systems design

  • Spend the session on Systems design, which Rad Ai candidates report being tested on.
  • Write one worked example in Systems design and time yourself on it.

Deliverable: One timed worked example in Systems design.

Day 5 — Answer out loud: Coding & Practical Implementation

  • Answer aloud, timed: Implement a robust data-parsing pipeline in TypeScript that processes asynchronous inputs and handles partial failures gracefully.
  • Answer aloud, timed: Create a reusable full-stack component that fetches, filters, and displays real-time data while optimizing for minimal network overhead.

Deliverable: Spoken answers to 2 reported Coding & Practical Implementation question(s), under time.

Day 6 — Answer out loud: System Design & Architecture

  • Answer aloud, timed: Design an asynchronous task-processing queue capable of handling high-throughput medical image analysis requests.
  • Answer aloud, timed: How would you structure a real-time collaborative editing platform for clinical reports, ensuring data consistency and low latency?

Deliverable: Spoken answers to 2 reported System Design & Architecture question(s), under time.

Day 7 — Answer out loud: Behavioral & Collaboration

  • Answer aloud, timed: Describe a time when you had to make a significant architectural tradeoff to meet a tight product deadline.
  • Answer aloud, timed: How do you handle a situation where a code reviewer strongly disagrees with your implementation choices?

Deliverable: Spoken answers to 2 reported Behavioral & Collaboration question(s), under time.

Behavioural preparation

Behavioural rounds judge the decision you made and what it cost.

Describe a time when you had to make a significant architectural tradeoff to meet a tight product deadline.

Medium · Behavioral & Collaboration

Approach

  1. Pick a story where you made the decision, not one where you watched it.
  2. State the situation in two sentences and spend the rest on the reasoning.
  3. Give the blast radius: what could have broken, and what you measured.
  4. Name the disagreement and how you resolved it with evidence.

Follow-up

  • What would you do differently if you ran that again?
  • How did you know your change caused the improvement?

How do you handle a situation where a code reviewer strongly disagrees with your implementation choices?

Medium · Behavioral & Collaboration

Approach

  1. Pick a story where you made the decision, not one where you watched it.
  2. State the situation in two sentences and spend the rest on the reasoning.
  3. Give the blast radius: what could have broken, and what you measured.
  4. Name the disagreement and how you resolved it with evidence.

Follow-up

  • What would you do differently if you ran that again?
  • How did you know your change caused the improvement?

Turn a code review disagreement into a decision

Easy · code review · optimistic concurrency · communication

A colleague's change updates a row with UPDATE resource SET version = version + 1 WHERE resource_id = $1 AND version = $2 and treats an affected-row count of zero as a successful no-op. You read that as a silently lost update; they think returning 200 is friendlier to clients than returning a conflict. Describe how you have handled a review disagreement of this shape: what goes in the comment, when you leave the thread, and who decides. Then write the comment you would leave here, in under 80 words.

Approach

  1. Sort the disagreement before writing anything. A silently discarded write is a correctness claim about data; the choice between 409 and 412 is taste. Only the first justifies blocking a merge, and saying which one you are doing is most of the value of the comment.
  2. Make the claim reproducible in the comment itself with an interleaving rather than a principle: A reads version 7, B reads version 7, B commits version 8, A's predicate matches zero rows, A is told it succeeded and A's edit is gone.
  3. Offer the alternative with its cost attached: return 409 carrying the current version and the revision that won, so the client can re-read and re-apply. Note that automatic retry is not the fix, because a retry re-reads the winner's state and reapplies an intent formed against data that no longer exists.
  4. Apply an escalation rule you can state: two round trips on the thread, then a call, and the service's owner decides rather than the reviewer. A reviewer who cannot be overruled is a bottleneck with extra steps.

Follow-up

  • Where would you put the test that fails if someone reintroduces the swallowed zero rowcount?
  • The author says clients cannot handle a 409. How do you check whether that is true?