Articles

Maki Assessment for Technical Candidates: Match Practice to Your Invitation

Prepare for a Maki assessment by matching practice to your invitation: reasoning, technical knowledge, spoken answers, timing rules, and score limits.

A Maki invitation is a preparation brief, not the name of one fixed exam. Before practicing algorithms, inspect the activities your employer actually selected. A technical candidate might need to reason about code, interpret numerical information, explain a decision on video, or complete another named activity. Preparing the wrong format can consume the time you need for the right one.

This guide separates official platform documentation, limited candidate reports, and original preparation exercises. It does not claim that every Maki assessment contains programming, video, or the same number of questions. The invitation and instructions shown for your assessment determine its scope.

What can you verify before choosing practice?

Official platform guidance: Maki’s preparation article describes an on-screen timer and question list, and says a tutorial may be available through a Trial run button. “May” matters: do not assume that every invitation supplies a sandbox. It also notes that assessment preparation depends on the test offered. Maki preparation guidance

Open the invitation and activity overview without launching a timed activity unnecessarily. Record the employer, role, deadline with time zone, activity names, response format, expected equipment, and any visible practice instructions. Distinguish information you actually saw from information you still need.

For example, “programming activity” is incomplete preparation evidence. Is it multiple choice, code-output prediction, an editor task, or a spoken explanation? A title mentioning Java does not tell you whether you must build a project. Likewise, “interview” does not establish whether the response is recorded once or collected through a conversation.

Use a simple rule: every practice choice should point back to an invitation field. If you cannot name that field, the exercise is general interview preparation. General practice can still be useful, but it should not displace an explicitly required activity.

Build an invitation-to-practice map

This is an original planning tool, not a list of mandatory Maki modules. Complete only rows corresponding to your invitation.

What the invitation actually says	A matching practice action	What to confirm before starting
Numerical reasoning or data interpretation	Read a small table; calculate rates and differences with units	Calculator policy, response format, and activity clock
Verbal reasoning	Classify statements using only the supplied passage	Whether answers distinguish false from unsupported
Logical or pattern reasoning	Translate constraints or sequences into explicit rules	Navigation and whether returning to an item is allowed
Technical knowledge or code questions	Predict output, identify an error, and explain one relevant concept	Named language, version, and whether code entry is required
Recorded video response	Deliver one concise example aloud and review the recording	Preparation time, response time, and attempt rules
Voice conversation	Answer a question directly and handle a follow-up	Whether the interaction is conversational and what setup it requires
Work preference or situational activity	Read the stated situation and answer consistently from your experience	What response scale or instructions apply

Suppose an original hypothetical invitation lists numerical reasoning, Java knowledge, and a recorded response. A sensible preparation session includes a rate table, short Java traces, and a spoken example. Spending the entire session on dynamic programming would miss two named formats. If another invitation lists only language and work preferences, the same Java session would be poorly matched.

Do not infer activity order, question counts, or a hiring cutoff from this map. Its purpose is to prevent a platform name from replacing the actual requirements.

What do candidate reports add—and where do they stop?

Candidate reports: in a public Wise graduate software-engineer discussion from the 2026 cycle, one participant described aptitude, programming multiple-choice questions, and a one-way response. Other participants discussed completing the assessment and waiting for further contact. These are individual accounts, not official documentation of your employer’s assessment. Wise graduate discussion

An older candidate account in a separate Wise thread described attention-to-detail tasks, technical knowledge questions, situational items, and video. That report includes uncertain recollections of counts and attempts; this guide does not turn them into present rules. Earlier Wise discussion

The research did not establish two independent current reports describing the same technical employer’s complete assessment. The scope is therefore deliberately narrower: how to match preparation to an invitation, rather than a guaranteed Wise or universal Maki sequence.

Editorial inference: the accounts are useful prompts to check whether “technical” means writing code or reasoning about code. They cannot tell you how long your activity runs, how your score is weighted, or which employer will invite you onward. A report that someone scored highly and waited is not evidence of a hidden passing threshold.

Practice numerical and technical reasoning on the same evidence

The following original exercise is not a Maki test item. It offers two response formats so you can practice the one your invitation requires.

A service dashboard shows:

Period	Requests attempted	Requests completed successfully
Before a release	200	180
After a release	250	200

For numerical practice, calculate both success rates and describe the change. The rates are 180/200 = 90% and 200/250 = 80%. Successful requests increased by 20, while the success rate fell by 10 percentage points. The relative decrease in the rate is approximately 10/90 = 11.1%.

Those are three different answers to three different questions. A hurried candidate might report a 10% decrease without distinguishing percentage points from relative percent. Another might say reliability improved because the absolute successful count increased. Read the requested metric before selecting an answer.

For technical practice, consider this pseudocode:

success_rate = completed_successfully / attempted

Ask what the language does when both operands are integers, what happens if attempted is zero, and whether the two counts refer to the same time window. State a denominator policy before patching the code. Do not change a missing denominator into a fabricated 100% success rate.

A robust explanation might say: “I would use the language’s intended floating-point or decimal arithmetic, validate that the denominator is positive, and keep the result distinct from an unavailable metric. I would test 180/200, 200/250, and zero attempts.” This is a reasoning exercise; the appropriate implementation depends on the language and contract.

For verbal practice, evaluate “the release caused the lower success rate.” The table shows an association across periods. It does not establish causality: traffic mix, retries, or another change could explain it. Practice separating an observed calculation from a stronger conclusion unsupported by the passage.

Prepare a spoken explanation without memorizing a script

Official video guidance: Maki describes answering a recruiter’s question by recording a clip, with an opportunity to configure and inspect camera and microphone settings before recording. It distinguishes setup from the running stopwatch. Maki video preparation

For independent practice, take the dashboard exercise and record a short explanation: what changed, what you can conclude, and what evidence you would request next. Review whether the opening answers the question, whether the numbers are accurate, and whether the recommendation follows the evidence. A polished introduction cannot compensate for confusing success counts with success rates.

For a behavioral prompt, select a real technical decision you can describe concretely. Explain your responsibility, the constraint, the action, the observed result, and what you changed afterward. Avoid attributing a team’s entire result to yourself. Prepare a few factual examples, not one rigid speech that you force into every question.

Official AI disclosure: Maki says some invitations can include a voice-based conversational interview, spoken-answer transcription, AI-assisted scoring, and automated monitoring. These features depend on assessment type; the disclosure describes human hiring decisions. A recorded clip and an AI-led conversation are different interactions. Maki AI disclosure

If the invitation specifies a conversation, practice answering follow-ups that change the question. If it specifies a recorded response, practice a complete answer within the stated response limit. Do not assume another candidate’s retry count applies to either format.

Which timing and technical rules deserve special attention?

Official candidate FAQ: Maki recommends current Chrome on a desktop or laptop. It says timed activities must be completed once started, while breaks between activities are possible within the assessment deadline. Its fullscreen warning rule applies when fullscreen is required. Technical-issue retakes are reviewed rather than automatically granted. Maki candidate FAQ

Convert these facts into a practical distinction: the expiration deadline is the last permissible completion boundary; an activity countdown limits an individual activity. Having hours before expiration does not mean you can pause a countdown for lunch. Leave enough room before the deadline to resolve a setup problem.

Before starting, close distracting notifications, verify the named input devices if needed, and read the rules about permitted materials. A recommendation to have a calculator available is not blanket permission to use one in every activity. Follow the instructions for that activity, including any rules on external tools or assistance.

If a problem occurs, record the activity name, approximate time, browser, visible error, and what action preceded it. Contact the appropriate support channel with a concise account. Do not repeatedly launch activities expecting a free reset or treat a practice example as permission to receive help during a live assessment.

What can a score or reused result tell you?

Official scoring information: Maki documents different methods for different activity types, including deterministic scoring and AI-supported evaluation. It describes language frameworks and structured behavioral grids, rather than one universal “percentage correct” interpretation. Maki scoring explanation

Official results information: sharing a report depends on the recruiter’s authorization. The report may contain an overall score, test-level scores, and dimensions. A missing report does not establish failure, and a received report does not establish an offer. Receiving results, Interpreting a report

Official reuse information: Maki’s recruiter documentation describes activity-result reuse with a default six-month validity period and possible activity-level variation. Treat an “already completed” message as a state to clarify, not evidence that the current invitation is broken or that you can reset it yourself. Candidate-profile reuse

For preparation, inspect a shared report at the activity level. A weak reasoning dimension suggests a different exercise from a weak spoken explanation. For application status, ask the employer; the assessment provider’s technical support does not replace the hiring team.

Choose your next exercise from the invitation

Write one sentence for each activity: “I need to answer in this format, under these stated constraints, using this skill.” Then choose a short independent exercise and review the mistakes it exposes.

Your goal is familiarity with the required task and accurate reasoning under its rules. A platform name, an impressive score posted online, or an older applicant’s question list cannot substitute for that preparation.