Directi Placement Papers: Test Pattern, Syllabus, and Sample Questions

Directi’s online test runs two consecutive elimination rounds, and the coding section is harder than the typical Indian campus placement test.

That gap between expectation and reality catches a lot of students. The company co-founded CodeChef, one of India’s primary competitive programming platforms, and the test difficulty reflects that orientation. Candidates who prepare only on general aptitude and basic coding questions find the coding round a step up.

This guide covers the full Directi online test: corporate context, test format, topic-wise syllabus for both rounds, and worked sample questions. For the interview rounds after the online test, see the Directi interview questions and rounds guide. For eligibility criteria and the full recruitment pipeline, see the Directi fresher recruitment process guide.

What Directi Is in 2026

Students searching for Directi placement papers sometimes get confused by conflicting results. The company has changed substantially since the 2016 to 2019 period when it was one of the more prominent campus recruiters in India.

Directi was founded in 1998 by Bhavin Turakhia and Divyank Turakhia in Mumbai as a diversified technology group. Several of its original business units were divested over the years:

Original Business Unit	What Happened	Current Status
Web hosting and domain units (ResellerClub, Webhosting.info)	Sold to Endurance International ~2014	Now part of Newfold Digital (rebranded 2021)
Media.net (contextual advertising)	Sold to a Chinese consortium in 2016 for ~$900M	Independent ad-tech company
CodeChef (competitive programming)	Acquired by Unacademy in 2020; spun off independently in 2023	Independent platform
Titan (email), Flock (comms), Radix (domains), Ringo (calling)	Still under Bhavin Turakhia’s entity umbrella	Active Directi-linked products

The “Directi” hiring that college placement cells reference today comes from the remaining product entities under the Directi/Bhavin Turakhia umbrella. Campus drives have been sparser since 2020 compared to the 2016 to 2019 peak period, but the company still hires engineers, primarily for its SaaS and infrastructure products.

The online test format described below has remained broadly stable across drives. Your placement cell will confirm whether a drive is scheduled at your college.

The Directi Online Test at a Glance

Directi’s online test has two rounds that function as back-to-back elimination filters. Clearing both is required to proceed to the interview stages.

Round	Questions	Duration	Marking
MCQ Round	20	30 minutes	+5 correct / −1 wrong / 0 unattempted
Coding Round	2–3 problems	90 minutes	Based on test cases passed

The MCQ round primarily targets Operations Engineer profiles; the coding round targets Software Engineer profiles. Some drives run both rounds in a single sitting; others schedule them on separate days. Confirm the format with your placement cell.

The coding round has historically been hosted on CodeChef’s platform. Some drives have also used HackerEarth. The platform used does not change the problem type or difficulty.

MCQ Round: Syllabus and Worked Sample Questions

The MCQ section tests core CS fundamentals across six topic areas. General aptitude typically accounts for two to four questions; the remaining questions are technical.

Topic Area	Typical Coverage
Operating Systems	Scheduling, memory management, file systems, the sticky bit, process states
Computer Networks	DNS resolution, IP addressing, OSI model, routing protocols
Data Structures	Trees, heaps, arrays, sorting algorithms
Algorithms	Time complexity, recursion, dynamic programming basics
DBMS	Normalisation, SQL, transactions, indexing
General Aptitude	Number systems, basic probability, series completion

The marking scheme rewards careful selection. With 20 questions and +5/−1 scoring, a wrong answer costs you the equivalent of one-fifth of a correct answer. Attempting 15 questions at high confidence outperforms attempting all 20 at moderate confidence.

MCQ Sample Questions

• Q1: The sticky bit is commonly set on which of the following directories in Linux?

  • a) /bin
  • b) /etc
  • c) /tmp
  • d) /var
  • Answer: c) /tmp. The sticky bit on a directory means only the file’s owner can delete or rename it, even if others have write permission on the directory. /tmp is the standard example: multiple users can create files there, but cannot delete each other’s files.

• Q2: A root DNS server responds to queries using which type of lookup?

  • a) Recursive
  • b) Iterative
  • c) Broadcast
  • d) Multicast
  • Answer: b) Iterative. Root DNS servers do not resolve queries on behalf of the client. They respond with a referral to the next-level authoritative server (the TLD server). The recursive resolution is handled by the resolver (typically the client’s ISP DNS server), not by the root itself.

• Q3: Which CPU scheduling algorithm is most likely to cause starvation?

  • a) Round Robin
  • b) First Come First Served
  • c) Priority Scheduling (without aging)
  • d) Shortest Job First (preemptive)
  • Answer: c) Priority Scheduling (without aging). When low-priority processes are perpetually passed over in favour of high-priority arrivals, they wait indefinitely. Round Robin avoids starvation by giving all processes equal time slices. Aging, which gradually increases a waiting process’s priority, is the standard fix for priority scheduling starvation.

• Q4: What is the time complexity of building a max-heap from an unsorted array of n elements?

  • a) O(n log n)
  • b) O(n)
  • c) O(log n)
  • d) O(n²)
  • Answer: b) O(n). The standard heapify-down (sift-down) approach, starting from the last non-leaf node and working up, achieves O(n) because lower levels of the heap require fewer operations. Inserting n elements one by one (sift-up) is O(n log n). The O(n) bound is why heap construction is more efficient than sorting by insertion.

Coding Round: Problem Types and Worked Example

The coding round contains two or three problems in 90 minutes. The difficulty sits above the standard campus aptitude test: expect problems that require algorithmic insight, not just syntax recall.

Problem categories that appear consistently across past Directi drives:

• Dynamic programming: 0/1 knapsack, maximum subarray, longest path in a tree
• Tree problems: maximum sum path between leaf nodes, vertical sum, product of leaf nodes
• Array manipulation: zig-zag rearrangement, maximum product subarray
• Graph traversal: shortest path with obstacles (BFS-based)
• Combinatorics: arrangements with constraints

The bonus problem in the three-problem format is typically harder than the other two and is scored separately. Solving the first two problems completely is more valuable than partially solving all three.

Practice resource: problems at difficulty level 3 and above on CodeChef closely match the difficulty and style of this round.

Worked Example: Array Zig-Zag Rearrangement

• Problem: Rearrange an array into a zig-zag pattern such that a[0] <= a[1] >= a[2] <= a[3] >= a[4] ...
• Example input: [4, 3, 7, 8, 6, 2, 1]
• Expected output: [3, 7, 4, 8, 2, 6, 1] (one valid arrangement)
• Step 1: Traverse the array. At every even index, the element should be less than or equal to its neighbours. At every odd index, it should be greater than or equal to its neighbours.
• Step 2: If the condition is violated, swap the adjacent pair.
  • Index 0: 4 should be <= arr[1]=3. Condition fails. Swap: [3, 4, ...]
  • Index 1: 4 should be >= arr[2]=7. Condition fails. Swap: [3, 7, 4, ...]
  • Index 2: 4 should be <= arr[3]=8. Condition holds. No swap.
  • Continue to end.
• Step 3: Time complexity is O(n) since each pair is visited once.
• Key insight: You do not need to sort the array. The single-pass swap approach works because each swap only affects the local condition and does not break the condition already satisfied for earlier positions.

Worked Example: Permutations (n people, r seats)

• Problem: In how many ways can n people occupy r seats in a theatre row?
• Formula: P(n,r) = n! / (n-r)!
• Example: n = 5 people, r = 3 seats.
  • P(5,3) = 5! / (5-3)! = 5! / 2! = (5 × 4 × 3 × 2 × 1) / (2 × 1) = 120 / 2 = 60 ways
• This is a direct application of the permutation formula. The 60 arrangements account for the different orderings of the 3 chosen people across the 3 seats.

Building Your Preparation Plan

The MCQ round and the coding round test different skill sets. Treating them as one undifferentiated task is the most common preparation mistake.

For the MCQ round (30 minutes, 20 questions):

• Revise OS scheduling algorithms, memory management, and the Linux file system model. The sticky bit and directory structure questions reward students who have covered their OS coursework.
• Refresh DNS resolution, IP address classification (private vs. public), and the OSI model for the networking questions. These are standard fare in the Cisco and Directi test formats alike — preparation for one transfers directly to the other.
• Practise mental calculation and series completion for the general aptitude questions. Two to four aptitude questions in 30 minutes do not require deep drilling, but cold starts hurt.

For the coding round (90 minutes, 2–3 problems):

• Solve 50 to 80 medium-level problems across dynamic programming, trees, and arrays before the drive. The D.E. Shaw placement guide covers the same difficulty band and problem categories — overlapping preparation is effective for both.
• Practise on CodeChef or HackerEarth specifically, since the Directi test is hosted on one of those platforms. Familiarity with the submission interface saves time on test day.
• Focus on getting two clean accepted solutions rather than attempting three at partial credit. Time allocation matters at this difficulty level.

The HirePro platform guide covers the test-administration patterns used across CodeChef-adjacent and HackerEarth-administered drives, which is useful context for understanding how the Directi test is structured and scored.

For the Cisco online test, the MCQ syllabus overlaps substantially with Directi’s: OS, networking fundamentals, and data structures. Students targeting both can build a shared preparation bank for the MCQ round.

The competitive programming depth this test expects also surfaces in AI systems work. Prompt routing, retrieval-augmented generation, and agent orchestration all draw on graph traversal and dynamic programming thinking at their core. If that intersection interests you, TinkerLLM at ₹299 is a low-commitment way to experiment with LLM applications before placement season is over; the skills are not separate tracks.