Advanced Coding Prompts: Architecture, Performance & Code Quality

Architecture decision

Act as a senior software architect. I need to decide between [option A] and [option B] for [the component or system]. Context: [current system state, scale, team size, language/stack]. My top concerns are [list 2–3 — e.g. maintainability, scalability, latency, team familiarity]. For each option: real tradeoffs (not just generic pros/cons), which scenarios favor it, and long-term maintenance implications. Recommend with rationale.

AI can surface tradeoffs you have not considered, but the final decision should factor in team context AI cannot know.

System design brainstorm

Act as a distributed systems engineer. I need to design [system description — e.g. a notification system / a rate limiter / a job queue / an analytics pipeline] that handles [scale context — e.g. 1M events/day / 10K concurrent users]. Walk me through: the core components, data flow, where state lives, failure modes and how to handle them, and the first implementation I should ship (not the final design).

Performance review

Act as a performance engineering expert. Review this [language] code for performance issues. Identify: time complexity of the main operations, memory usage concerns, unnecessary allocations or copies, any N+1 patterns, and opportunities to parallelize or cache. Prioritize findings by expected impact. Code: [paste].

Security code review

Act as a security engineer. Review this [language] code for security vulnerabilities. Check for: injection vulnerabilities (SQL, command, LDAP), improper input validation, authentication and authorization issues, sensitive data exposure, insecure dependencies or patterns, and any logic that could be exploited by an adversary. Report findings with severity (critical / high / medium / low) and suggested remediation. Code: [paste].

API design review

Act as a senior API designer. Review this API design for: consistency in naming conventions, appropriate use of HTTP methods and status codes, error response structure, versioning strategy, authentication approach, and any resource modeling issues. Then give 3 concrete improvements. API spec or endpoint list: [paste].

Refactoring plan

Act as a senior developer. I have a [language] codebase that [describe the problem — is too coupled / has unclear boundaries / is difficult to test / has grown past its original design]. Give me a step-by-step refactoring plan that minimizes risk and can be done incrementally without breaking the system. Start with the highest-leverage change. Do not rewrite everything — prioritize.

Database schema review

Act as a database architect. Review this schema for: normalization issues, missing indexes for likely query patterns, inappropriate data types, missing constraints, and any design choices that will hurt performance at scale. Describe the query patterns you would expect for [this application type], then evaluate how well the schema supports them. Schema: [paste DDL or description].

Debugging a complex issue

I have an intermittent bug in [language / system] that is hard to reproduce. Symptoms: [describe what happens, when, and how often]. Context: [describe system, environment, recent changes]. I have already tried: [list what you have ruled out]. Help me build a systematic debugging strategy — what to instrument, what logs to capture, what conditions to isolate, and what my leading hypothesis should be based on these symptoms.

Code review feedback framing

Act as a senior engineer doing a constructive code review. I am reviewing a PR from a junior developer. Here is the code: [paste]. Write a code review comment that: identifies the specific issue clearly, explains the 'why' behind the concern (not just the 'what'), suggests a concrete improvement, and is phrased in a way that teaches rather than criticizes. Focus on the most important finding.

Technical debt triage

Act as a software architect. I have accumulated technical debt in [area of codebase]. Here is a description of the issues: [describe 3–5 debt items]. Help me triage this backlog: categorize each item by type (design debt / code debt / test debt / documentation debt), estimate relative risk of not addressing each, and suggest a prioritization order based on [my primary concern: shipping velocity / stability / onboarding new engineers].

Build your own prompts

Use AI Prompt Generator to generate structured prompts tailored to your specific task, tone, and audience.

Open AI Prompt Generator