AMD Software Engineer interviews reported across candidates combine project discussion, coding, systems or hardware-adjacent topics, and behavioral conversations. Format and depth vary substantially by team.
$160K
Avg. Base Comp
$205K
Avg. Total Comp
2-5 rounds
Typical Rounds
2-4 weeks
Process Length
AMD Software Engineer interview reports show that the technical emphasis can vary considerably by team. Be ready to explain your reasoning and engineering decisions in detail. One ML-oriented candidate encountered a two-round process consisting of a technical interview and an HR conversation. The technical portion covered Linux process data, Python scripting, and both ML theory and code. Practice reading raw output from commands such as ps, identifying fields such as process memory use, and explaining how you extract the requested information.
A separate candidate described five hour-long technical onsite interviews. That process included navigating an open-source repository, solving a basic hash-map problem, discussing system and API design across Windows, Linux, and Mac, and debugging and explaining existing code. Resume and project discussion also appeared throughout, and one interviewer asked for a detailed explanation of transformers. This candidate found the debugging and code-explanation work more difficult than the algorithmic exercise.
These accounts support preparing beyond standard coding questions. Review every project on your resume so you can explain your contribution and technical choices. Practice exploring an unfamiliar repository aloud, tracing code behavior, diagnosing defects, and discussing design tradeoffs. For roles connected to AI or ML, retain precise ML knowledge while also reviewing Linux and Python fundamentals. The reported formats differ sharply, so use the job description and team domain to decide where to place the greatest preparation emphasis rather than assuming one universal AMD sequence.
Synthesized from 13 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Amd process.
The process consisted of three rounds: an HR screening call, a technical interview, and a final panel discussion with engineers and the hiring manager.
The HR round focused on my background, projects, current role, and motivation for switching. The technical round was much more project-driven than I expected. The interviewer spent nearly 20 minutes discussing one of my projects and kept drilling deeper into architectural decisions.
One specific question I was asked was:
"Suppose your application grows from 10,000 users to 1 million users. What would be the first three bottlenecks you would expect, and how would you address them?"
I discussed database indexing, caching with Redis, horizontal scaling of backend services, and CDN usage for static assets. This led to several follow-up questions around PostgreSQL performance and API optimization.
I was also given a coding problem involving finding the first non-repeating character in a string and then discussing the time and space complexity of my solution. The interviewer was more interested in my thought process and tradeoffs than simply getting the correct answer.
The final panel was the most challenging part because multiple interviewers focused on different areas. One interviewer concentrated on backend architecture, another on databases and cloud infrastructure, and another on system design. A question that caught me off guard was:
"Why did you choose PostgreSQL over MongoDB for this project, and under what circumstances would you switch?"
I felt most confident discussing projects that I had personally built because I could explain not only the implementation but also the reasoning behind each technical decision. The toughest moments came when interviewers kept asking "why" several levels deeper than I expected.
Overall, the interview felt less like a test of memorized knowledge and more like a discussion about real engineering decisions, tradeoffs, and problem-solving in production systems.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Amd
How would you build and justify the components of a Transformer encoder layer in PyTorch for large-scale text data?
| Question | |
|---|---|
| Relational Migration | |
| LRU Cache 1 | |
| 2nd Highest Salary | |
| Merge Sorted Lists | |
| Prime to N | |
| Find the Missing Number | |
| Random SQL Sample | |
| Maximum Profit | |
| Upsell Transactions | |
| Over-Budget Projects | |
| The Brackets Problem | |
| Hurdles In Data Projects | |
| Recurring Character | |
| Detecting ECG Tachycardia Runs | |
| Cyclic Detection | |
| Retailer Data Warehouse | |
| Employee Project Budgets | |
| Integer to Roman | |
| Sum to N | |
| Size of Joins | |
| Twenty Variants | |
| Equivalent Index | |
| Bagging vs Boosting | |
| Get Top N Frequent Words | |
| Completed Shipments | |
| Delivery Estimate Model | |
| One Element Removed | |
| Paired Products | |
| Groups of Anagrams |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report an initial conversation that may cover their background, projects, interest in AMD, current role, and motivation. Some reports describe an HR round, while others begin directly with technical discussion, so prepare a concise explanation of role fit.
Several candidates report detailed follow-up on internship or project work. Be ready to explain the architecture, choices you made, tradeoffs, and what you personally implemented; interviewers may continue asking why at progressively deeper levels.
Technical exercises reported include hash-map and string problems, tree questions, matrix multiplication, debugging, and explaining unfamiliar code. Candidates describe discussion of complexity and reasoning, so communicate assumptions and tradeoffs while working.
Candidates report team-dependent topics including Linux process output, Python scripting, ML theory, database and API scaling, timing analysis, and cross-platform design. Typically, the most relevant preparation follows the role’s technical domain and your stated experience.
Some candidates report a final HR round, while others describe a panel with engineers and a hiring manager. Expect this stage to potentially combine behavioral questions with further project, design, or technical follow-up rather than a uniform closing format.
