As a Software Engineer at Savvy, you are not just writing code; you are a product-minded builder tasked with integrating Applied AI directly into the advisor platform. This role is central to Savvy’s mission of automating complex financial workflows, allowing advisors to focus on client growth rather than manual operations. You will operate at the core of the product flywheel, designing and shipping intelligent agents and AI-powered systems that have a direct, measurable impact on the business. This position is designed for engineers who thrive in high-ownership, fast-moving environments. You will work across the full product development lifecycle—from identifying opportunities for automation to shipping features and iterating based on real-time feedback from financial advisors. If you are a systems-oriented engineer who enjoys navigating ambiguity and connecting technical implementation directly to user outcomes, this role offers a rare opportunity to shape the future of financial technology. 02 · The loop
Initial Screening
reportedThe first step involves a review of the candidate's application and background.
What to demonstrate
- The first step involves a review of the candidate's application and background
- Depth in Production AI Systems
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.
Technical Assessment
reportedCandidates undergo a technical evaluation to assess their coding skills and problem-solving abilities.
What to demonstrate
- Candidates undergo a technical evaluation to assess their coding skills and problem-solving abilities
- Depth in Production AI Systems
How to prepare
- Answer aloud and timed: How do you approach state management in a React and Next.js application?
- Answer aloud and timed: When designing a database schema in PostgreSQL, how do you balance performance with future scalability?
Leadership Discussions
reportedFinal discussions with leadership to evaluate cultural fit and alignment with company values.
What to demonstrate
- Final discussions with leadership to evaluate cultural fit and alignment with company values
- Depth in Production AI Systems
How to prepare
- Answer aloud and timed: Describe your experience integrating third-party AI services into a production backend.
- Answer aloud and timed: How would you architect a system to generate automated meeting notes from unstructured client data?
PracHub editorial advice for the preparation topics above.
Focus on the "Why
When explaining your technical decisions, always connect them back to the user or business outcome.
Be Ready to Demo
Since we value builders, be prepared to talk about specific features you have shipped in the past and how you measured their success.
Embrace Ambiguity
If an interview question feels open-ended, ask clarifying questions. We want to see how you structure your thoughts in the face of uncertainty.
Show Your AI Fluency
Don't just say you use AI; explain how you integrate it into your development lifecycle to increase your personal velocity.
Choose a category, try a prompt, then open its approach, worked solution or follow-up when you need it.
How have you used direct customer feedback to iterate on a feature you previously shipped?
How have you used direct customer feedback to iterate on a feature you previously shipped?
Approach
- Restate the input: its shape, its size, and what is guaranteed about it.
- Name the brute-force solution and its complexity before improving on it.
- Choose the data structure from the access pattern, not from familiarity.
- 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?
Track a rolling failure rate per destination for circuit decisions
The egress service delivers about 1,500 webhooks per second across roughly 40,000 destinations, each call bounded by a 10 second timeout. Maintain, per destination, the failure rate over the trailing 60 seconds so a caller can ask before dispatch whether the circuit should open. Attempts arrive as (destination_id, finished_at_ms, outcome). Requirement: amortised O(1) per attempt, with total memory bounded by the destination count rather than by traffic. Give the structure, its exact memory, and the rule that stops a destination with three attempts from opening a circuit.
Approach
- Name the exact-deque version and then reject it as the default. Holding timestamps and advancing a tail pointer past anything older than now minus 60 seconds is a correct two-pointer window at amortised O(1) per attempt, but its memory tracks in-window traffic, so one destination in a retry storm holds hundreds of thousands of entries while thousands of quiet destinations hold none.
- Use a ring of 60 one-second buckets per destination, each bucket a pair of counters for attempts and failures. On an attempt, advance the ring by the elapsed whole seconds, zeroing at most min(elapsed, 60) buckets, then increment the head. That is amortised O(1) with a fixed footprint per destination.
- State the footprint: 60 buckets times two 4-byte counters is 480 bytes of payload per destination, so 40,000 destinations is roughly 20 to 25 MB with per-entry overhead, bounded by the catalogue rather than by the rate. The cost is granularity, since the oldest bucket ages out in whole seconds, which is far tighter than the decision needs.
- Require a minimum sample before the circuit may open. A destination with three attempts and three failures reads as 100 percent and is not evidence; a floor of roughly 20 attempts in the window makes the ratio meaningful, and below that floor use a run of consecutive failures as the trigger instead.
Follow-up
- The fleet is 30 instances and each sees roughly a thirtieth of a destination's traffic. Where does the rate actually live, and what does a per-instance answer get wrong?
- A destination answers in 9.5 seconds and succeeds. It is not failing but it is consuming your per-destination concurrency. What signal should open the circuit here?
Canonicalise a request body into a stable idempotency fingerprint
idempotency_key.request_fingerprint is a SHA-256 over the method, path and canonicalised body, and a retry whose fingerprint differs must be rejected with 422 rather than served the stored response. Write the canonicaliser. Bodies are JSON up to 256 KB nested at most 32 levels; clients vary key order, whitespace and unicode escaping, and some send 64-bit ids as JSON numbers. Produce a deterministic byte string such that semantically identical bodies match and any semantic difference does not. State your complexity and name two normalisations you refuse to perform.
Approach
- Parse once into a tree, then re-serialise under fixed rules: object keys sorted, array order preserved, one escaping convention, no insignificant whitespace. Parsing is O(n) and sorting keys is O(k log k) per object, so O(n log n) overall with O(depth) stack, and the 32-level cap is enforced during parsing because hostile nesting is how a canonicaliser becomes a stack overflow.
- Sort keys by their UTF-8 bytes and say why the obvious implementation is wrong in some runtimes: a default string comparison that orders by UTF-16 code units places surrogate pairs, meaning code points from U+10000 up, below U+E000 to U+FFFF, which is not UTF-8 byte order, so two services written in different languages disagree on the same document.
- Do not re-encode numbers through a double. IEEE-754 binary64 represents integers exactly only up to 2^53, so normalising a 19-digit id through a float changes it, and 1 against 1.0 cannot be reconciled without deciding whether they are the same value. Preserve the literal token, and require ids as strings at the API boundary if you want them comparable.
- Reject duplicate keys rather than picking one. JSON permits them and parsers disagree, most keeping the last, so any choice you make ties the fingerprint to a parser detail that the code handling the request does not necessarily share.
Follow-up
- A client sends the same logical request with an extra field your API ignores. Same key, different fingerprint, so you return 422. Is that the right answer?
- Where does the fingerprint get computed relative to request decompression and the body-size limit?
Keep soft-deleted accounts from blocking re-registration
app_user holds user_id, tenant_id, email CITEXT, password_hash (NULL for SSO principals), email_verified_at, auth_version, status ('invited','active','suspended','deactivated'), created_at, updated_at, deleted_at. Two live accounts for one address inside a tenant must be impossible, but an address freed by a soft delete must be reusable, and the same tenant may delete and re-register it repeatedly. Write the uniqueness DDL for PostgreSQL 16, then the equivalent for MySQL 8 where partial indexes do not exist, and say what each permits once three deleted rows already hold that address.
Approach
- Start from what is actually unique: not (tenant_id, email), but (tenant_id, email) among live rows. PostgreSQL says that directly — CREATE UNIQUE INDEX app_user_live_email ON app_user (tenant_id, email) WHERE deleted_at IS NULL. A full constraint over the same two columns burns the address permanently the first time someone deletes an account.
- Keep case-insensitivity in the type or the index, never in the application: CITEXT as given, or UNIQUE (tenant_id, lower(email)) as an expression index where the extension is unavailable. A case-sensitive unique column is exactly how two accounts for one human appear.
- For MySQL 8 the predicate has to move inside the key: add a discriminator column that is a constant 0 while the row is live and is set to user_id on delete, with UNIQUE (tenant_id, email, deleted_marker). Live rows share the constant and still collide; deleted rows differ from each other and stop colliding.
- State the NULL variant and its dependency: leaving the marker NULL for deleted rows also works, because a unique index treats NULLs as distinct — true in MySQL, and true in PostgreSQL only under the default NULLS DISTINCT, which PostgreSQL 15 lets you reverse. Check the polarity against the three existing deleted rows: constant-on-live is what preserves the collision you want, and reversing it silently admits duplicate live accounts.
Follow-up
- A deleted account re-registers with the same address the next day. Do the old resource rows follow the new user_id, and how does the API keep the two principals apart?
- How do you honour an erasure request while resource_revision.actor_user_id still references this table?
Replace offset paging on the resource feed with keyset
resource holds resource_id, tenant_id, owner_user_id, title, body_ref, version, status ('draft','active','archived','deleted'), created_at, updated_at, deleted_at, with an index on (tenant_id, status, updated_at DESC, resource_id DESC). The listing endpoint returns active resources for one tenant, newest update first, 50 per page, today with LIMIT 50 OFFSET n. Tenants reach page 400 and rows are created while they read. Write the keyset query, define what the cursor carries and how it is encoded, and say which part of the index each predicate uses. Assume PostgreSQL 16.
Approach
- Name the two failures separately. OFFSET 20000 makes the server produce and discard 20,000 rows, so page cost grows with depth rather than with page size. Independently, any write that changes how many rows sort above the offset moves the window between two fetches, and the direction decides which anomaly you get: an insert lands at the head of updated_at DESC and pushes already-returned rows down past the boundary, so they are returned a second time; a delete above the offset, or a row whose updated_at is bumped above the cursor, pulls rows up and one is never returned at all. Nothing in the response reveals either.
- Write the seek: WHERE tenant_id = $1 AND status = 'active' AND (updated_at, resource_id) < ($2, $3) ORDER BY updated_at DESC, resource_id DESC LIMIT 50. The row-value comparison is one index range rather than a disjunction, and both columns are NOT NULL, which is what makes that comparison well defined.
- Map each predicate onto the index: tenant_id and status are equality on the leading columns, (updated_at, resource_id) is the range, and the ORDER BY matches the index order so no Sort node appears and the scan stops after 50 rows. The DESC in the definition only matters for mixed directions — a plain ascending btree on the same columns is read backwards for this query.
- Put both sort columns in the cursor and nothing the client can tamper with into another tenant: base64 of (updated_at, resource_id), validated server-side, with tenant_id taken from the principal.
Follow-up
- The client asks for 'jump to page 400'. What do you offer instead, and what does the honest version cost?
- Sort order becomes user-selectable across four columns. How many indexes is that, and which would you refuse to add?
How would you design an API to handle real-time updates for an advisor dashboard?
How would you design an API to handle real-time updates for an advisor dashboard?
Approach
- Fix the scope first: who calls this, how often, and what they do when it fails.
- Name the read and write paths separately; they rarely have the same bottleneck.
- Choose a partition key and say what query it makes expensive.
- 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?
Can you explain a complex bug you encountered in a Ruby on Rails or TypeScript environment and how you resolve
Can you explain a complex bug you encountered in a Ruby on Rails or TypeScript environment and how you resolved it?
Approach
- Clarify what is being asked and what a complete answer contains.
- State your assumptions explicitly before working the problem.
- Say what you would check first and why it is the highest-information step.
- Work from the requirement backwards to the design.
Follow-up
- What assumption would you test first?
- How would you know your answer was wrong?
How do you approach state management in a React and Next.js application?
How do you approach state management in a React and Next.js application?
Approach
- Clarify what is being asked and what a complete answer contains.
- State your assumptions explicitly before working the problem.
- Say what you would check first and why it is the highest-information step.
- Work from the requirement backwards to the design.
Follow-up
- What assumption would you test first?
- How would you know your answer was wrong?
When designing a database schema in PostgreSQL, how do you balance performance with future scalability?
When designing a database schema in PostgreSQL, how do you balance performance with future scalability?
Approach
- Fix the scope first: who calls this, how often, and what they do when it fails.
- Name the read and write paths separately; they rarely have the same bottleneck.
- Choose a partition key and say what query it makes expensive.
- 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 generate automated meeting notes from unstructured client data?
How would you architect a system to generate automated meeting notes from unstructured client data?
Approach
- Fix the scope first: who calls this, how often, and what they do when it fails.
- Name the read and write paths separately; they rarely have the same bottleneck.
- Choose a partition key and say what query it makes expensive.
- 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?
Explain how you would implement a RAG (Retrieval-Augmented Generation) pipeline for an advisor-facing assistan
Explain how you would implement a RAG (Retrieval-Augmented Generation) pipeline for an advisor-facing assistant.
Approach
- Say who the caller is and what they do when the call fails halfway.
- Define the identity of a request so a retry cannot double-apply it.
- Separate accepted, pending, failed and confirmed; they are different facts.
- 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?
What strategies do you use to ensure the reliability and latency of AI agents in a production environment?
What strategies do you use to ensure the reliability and latency of AI agents in a production environment?
Approach
- Say who the caller is and what they do when the call fails halfway.
- Define the identity of a request so a retry cannot double-apply it.
- Separate accepted, pending, failed and confirmed; they are different facts.
- 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?
One log partition stops advancing while the others drain
Search results for a subset of tenants are hours stale; the rest are current. The projection consumer reports lag of zero on 15 of 16 partitions and 400,000 on one. Its error rate is flat and its CPU is idle. outbox_event has no pending rows older than a second, so the relay has published everything it holds. Identify the mechanism, give the ordered checks, and state what you do in the first ten minutes versus what you change permanently.
Approach
- Read the lag distribution first. A slow consumer lags everywhere; zero on fifteen partitions and 400,000 on one is not throughput. Idle CPU on the stuck partition means the consumer is not advancing its offset at all, which points at one message it cannot get past rather than at a rate problem.
- Exonerate the producer before touching the consumer. No pending outbox rows older than a second means the relay published, so the event exists in the log. This separates never sent from sent and never applied, which are different code paths and usually different owners.
- Read the message at the stuck offset and the handler's log lines for its event_id. A flat error rate with no progress has two explanations and you must distinguish them: the handler is throwing and the retry loop is swallowing it, or the handler is blocking on something and never returning. Idle CPU with no error lines favours the second.
- Mitigate before diagnosing further. Move the offending event to a dead-letter store and commit the offset past it. Adding consumers does nothing here, because a partition is consumed by exactly one member of the group, and the blast radius is every aggregate hashed to that partition, not only the aggregate that produced the bad event.
Follow-up
- The dead-lettered event carried aggregate_version 7 and the projection had applied 6. What must the replay do differently if 8 and 9 landed in the meantime?
- How do you show staleness to the user while the partition is behind, given the API already returns the projection's watermark?
Built from the rounds and topics Savvy candidates report.
Prepare, practise & reflect
One practical outcome each day. Spend longer where you need it.
0 / 7 done01Map the Savvy loop
- Write out the reported sequence: Initial Screening, Technical Assessment, Leadership Discussions.
- 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 3 reported rounds, with the weakest marked.
02Work Production AI Systems
- Spend the session on Production AI Systems, which Savvy candidates report being tested on.
- Write one worked example in Production AI Systems and time yourself on it.
Deliverable: One timed worked example in Production AI Systems.
03Work Ruby on Rails
- Spend the session on Ruby on Rails, which Savvy candidates report being tested on.
- Write one worked example in Ruby on Rails and time yourself on it.
Deliverable: One timed worked example in Ruby on Rails.
04Work AI Agents
- Spend the session on AI Agents, which Savvy candidates report being tested on.
- Write one worked example in AI Agents and time yourself on it.
Deliverable: One timed worked example in AI Agents.
05Answer out loud: Technical & Domain Proficiency
- Answer aloud, timed: How would you design an API to handle real-time updates for an advisor dashboard?
- Answer aloud, timed: Can you explain a complex bug you encountered in a Ruby on Rails or TypeScript environment and how you resolved it?
Deliverable: Spoken answers to 2 reported Technical & Domain Proficiency question(s), under time.
06Answer out loud: System Design & AI Integration
- Answer aloud, timed: How would you architect a system to generate automated meeting notes from unstructured client data?
- Answer aloud, timed: Explain how you would implement a RAG (Retrieval-Augmented Generation) pipeline for an advisor-facing assistant.
Deliverable: Spoken answers to 2 reported System Design & AI Integration question(s), under time.
07Answer out loud: Behavioral & Product Ownership
- Answer aloud, timed: Tell me about a time you had to move quickly with limited information; how did you define success?
- Answer aloud, timed: How do you prioritize technical debt versus shipping new features for users?
Deliverable: Spoken answers to 2 reported Behavioral & Product Ownership question(s), under time.
Expand any day for tasks and deliverables. Your progress is saved on this device.
Behavioural rounds judge the decision you made and what it cost.
Describe your experience integrating third-party AI services into a production backend.
Describe your experience integrating third-party AI services into a production backend.
Approach
- Pick a story where you made the decision, not one where you watched it.
- State the situation in two sentences and spend the rest on the reasoning.
- Give the blast radius: what could have broken, and what you measured.
- 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 failure states in a multi-step automated workflow?
How do you handle failure states in a multi-step automated workflow?
Approach
- Pick a story where you made the decision, not one where you watched it.
- State the situation in two sentences and spend the rest on the reasoning.
- Give the blast radius: what could have broken, and what you measured.
- 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?
Tell me about a time you had to move quickly with limited information; how did you define success?
Tell me about a time you had to move quickly with limited information; how did you define success?
Approach
- Pick a story where you made the decision, not one where you watched it.
- State the situation in two sentences and spend the rest on the reasoning.
- Give the blast radius: what could have broken, and what you measured.
- 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 prioritize technical debt versus shipping new features for users?
How do you prioritize technical debt versus shipping new features for users?
Approach
- Pick a story where you made the decision, not one where you watched it.
- State the situation in two sentences and spend the rest on the reasoning.
- Give the blast radius: what could have broken, and what you measured.
- 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?
Describe a situation where you had to translate complex technical constraints to a non-technical stakeholder.
Describe a situation where you had to translate complex technical constraints to a non-technical stakeholder.
Approach
- Pick a story where you made the decision, not one where you watched it.
- State the situation in two sentences and spend the rest on the reasoning.
- Give the blast radius: what could have broken, and what you measured.
- 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?
- 01
Describe your experience integrating third-party AI services into a production backend.
- 02
How do you handle failure states in a multi-step automated workflow?
- 03
Tell me about a time you had to move quickly with limited information; how did you define success?
- 04
How do you prioritize technical debt versus shipping new features for users?
How difficult are the technical interviews?
The technical interviews are rigorous but practical. We focus on real-world engineering problems rather than abstract puzzles, so be prepared to discuss your past projects and architectural choices in depth.
Savvy Software Engineer candidate reports ↗What is the company culture like?
Savvy is a fast-growing, high-velocity startup. We value autonomy, direct communication, and a deep focus on the user experience. You will be expected to own your work and contribute to the product roadmap.
Savvy Software Engineer candidate reports ↗How long does the interview process typically take?
While it can vary based on the team and candidate availability, we aim to be efficient. Most candidates move through the process in a few weeks, starting with a screen and ending with a conversation with leadership.
Savvy Software Engineer candidate reports ↗Is remote work an option?
We operate in a high-ownership environment that often requires close collaboration. Specifics regarding remote or hybrid work are discussed during the initial screening phase based on the team's needs.
Savvy Software Engineer candidate reports ↗What topics does Savvy test in interviews?
Savvy interviews most often cover Production AI Systems, Ruby on Rails, AI Agents, Workflow Automation, and PostgreSQL. The exact emphasis depends on the specific role you apply for.
Savvy Software Engineer candidate reports ↗Sources & methodology 3 sources ↗
No official company page is cited. Rounds and questions come from candidate reports and PracHub editorial material; each source shows the date it was read.
- 01Savvy Software Engineer candidate reports ↗
Company-reported rounds, questions and FAQ.
Candidate reports · Accessed 2026-09-22 - 02PracHub Software Engineer practice ↗
PracHub practice material, not company-reported.
PracHub page · Accessed 2026-09-22 - 03PracHub preparation framework ↗
PracHub preparation guidance.
PracHub page · Accessed 2026-09-22
Explore related engineering environments and compare the preparation focus.
Discord Software Engineer Interview Guide
Database practice for Discord candidates: how an index changes a query plan, and which anomalies survive a given isolation level, since read committed still permits non-repeatable reads.
Rystad Energy Software Engineer Interview Guide
Prepare for Rystad Energy software interviews with worked Python validation, OOP and decorator examples, plus an ingestion case and practical checks.
Zwift Software Engineer Interview Guide
What Zwift asks Software Engineer candidates, from reported rounds and questions.
Zumper Software Engineer Interview Guide
What Zumper asks Software Engineer candidates, from reported rounds and questions.