{"slug": "akuna-capital-oa-checks-whether-you-used-ai-even-analyzing-your-syntax-tree", "title": "Akuna Capital OA Checks Whether You Used AI — Even Analyzing Your Syntax Tree", "summary": "A developer reported that Akuna Capital's online assessment on HackerRank includes advanced AI detection methods, such as analyzing the code's abstract syntax tree and comparing patterns against typical AI-generated solutions. The assessment also features AI trap questions designed to mislead AI reasoning. The developer noted that the question pool is small with high overlap, making pattern review effective for preparation.", "body_md": "I just finished Akuna Capital's OA on HackerRank. The format was three coding problems in two hours.\n\nThe problems themselves were not extremely difficult, but there is one thing worth mentioning first.\n\nAkuna Capital clearly states in the OA instructions that AI tools are not allowed.\n\nTheir detection methods are also quite advanced: they analyze the code's abstract syntax tree (AST),\n\ncompare patterns against typical AI-generated solutions, and check for similarities between candidates'\n\nsubmissions.\n\nEven more interestingly, they have reportedly included AI trap questions.\n\nSome problem statements contain details designed to mislead AI reasoning, causing generated solutions\n\nto fail in specific ways. For candidates who rely heavily on vibe coding, this OA is definitely not\n\nsomething to gamble on.\n\nThe good news is that Akuna's question pool is relatively small. There are only around a dozen recurring\n\nproblems, and the overlap rate is quite high. Here is a breakdown of the three questions I encountered.\n\nA store wants to arrange products from highest popularity to lowest popularity. Each product has a unique\n\npopularity value. In one operation, you can swap any two products. Find the minimum number of swaps needed.\n\nExample:\n\n`[3, 4, 1, 2]`\n\nSwap 3 and 4:\n\n`[4, 3, 1, 2]`\n\nSwap 1 and 2:\n\n`[4, 3, 2, 1]`\n\nThe answer is 2 swaps.\n\nThis is a classic permutation cycle problem.\n\nMap every element's current position to its correct position after sorting.\n\nThe array can then be divided into several cycles.\n\nA cycle of length `k`\n\nrequires exactly `k - 1`\n\nswaps to fix.\n\nTherefore, the answer is:\n\n**number of elements - number of cycles**\n\nImplementation steps:\n\nA company delivers packages between cities. Cities are connected by bidirectional roads, and some cities\n\nmay not be reachable.\n\nDelivery order follows these rules:\n\nExample:\n\nFour cities with roads:\n\n`1-2, 2-3, 2-4`\n\nStarting from city 1:\n\n`[2, 3, 4]`\n\nUse BFS to calculate the shortest distance from the starting city because all edges have equal weight.\n\nThen sort reachable cities by:\n\n`(distance, city_number)`\n\nImportant edge cases:\n\nAlex has a sequence of movie ratings, including positive and negative values.\n\nHe wants to select a subsequence with the maximum possible total rating.\n\nHowever, he cannot skip two movies consecutively.\n\nExample:\n\nFor:\n\n`[-1, -3, -2]`\n\nThe second movie must be selected, or both the first and third movies must be selected.\n\nThe maximum score is:\n\n`-3`\n\nAnother example:\n\n`[-3, 2, 4, -1, -2, -5]`\n\nThe optimal selection is:\n\n`[2, 4, -2]`\n\nwith a total score of 4.\n\nThis is a linear dynamic programming problem.\n\nThe rule \"cannot skip two consecutive movies\" means the distance between any two selected indexes\n\ncannot exceed 2.\n\nDefine:\n\n`dp[i]`\n\n= maximum score when selecting the i-th movie.\n\nTransition:\n\n`dp[i] = ratings[i] + max(dp[i-1], dp[i-2])`\n\nThe final answer is:\n\n`max(dp[n-1], dp[n-2])`\n\nbecause skipping the last movie is allowed, but skipping the last two movies is not.\n\nThese three problems test three classic patterns:\n\nThe common theme is that the problems are wrapped in business scenarios:\n\nproduct ranking, delivery systems, and movie festivals.\n\nOnce you remove the story, they are standard algorithm patterns.\n\nThe most valuable information is that Akuna's OA question pool is relatively small and has a high overlap\n\nrate. Reviewing the recurring patterns before the assessment can significantly improve your preparation\n\nefficiency.\n\nI collected these verified Akuna OA questions and similar quantitative interview materials through\n\n[InterviewShow](https://interviewshow.com/).\n\nFor companies with smaller question pools, practicing representative problems before the OA usually\n\nprovides the highest return.\n\nGood luck with your interviews!", "url": "https://wpnews.pro/news/akuna-capital-oa-checks-whether-you-used-ai-even-analyzing-your-syntax-tree", "canonical_source": "https://dev.to/interviewshow-cs/akuna-capital-oa-checks-whether-you-used-ai-even-analyzing-your-syntax-tree-10a2", "published_at": "2026-07-16 12:00:36+00:00", "updated_at": "2026-07-16 12:34:07.773280+00:00", "lang": "en", "topics": ["developer-tools", "artificial-intelligence"], "entities": ["Akuna Capital", "HackerRank", "InterviewShow"], "alternates": {"html": "https://wpnews.pro/news/akuna-capital-oa-checks-whether-you-used-ai-even-analyzing-your-syntax-tree", "markdown": "https://wpnews.pro/news/akuna-capital-oa-checks-whether-you-used-ai-even-analyzing-your-syntax-tree.md", "text": "https://wpnews.pro/news/akuna-capital-oa-checks-whether-you-used-ai-even-analyzing-your-syntax-tree.txt", "jsonld": "https://wpnews.pro/news/akuna-capital-oa-checks-whether-you-used-ai-even-analyzing-your-syntax-tree.jsonld"}}