BuildAdvisor AI uses local data and rule-based engineering checks before it recommends a PC build. The goal is not to sound smart—it is to make the decision process visible, understandable, and repeatable.
Engineering flow
A candidate must pass compatibility filters before the scoring system can rank it.
Budget, use case, resolution, priorities
Remove incompatible combinations
Rank valid builds with local rules
Show tradeoffs, warnings, and upgrades
Under the hood
The questionnaire turns a beginner’s goals into structured inputs: budget, main use, games, display resolution, priorities, Wi-Fi needs, compact-case preference, peripherals, and any parts they already own. The engine uses these answers as constraints, not as prompts for a random parts list.
Every component lives in a typed local TypeScript catalog. Samples include static price estimates, CPU core counts, GPU memory, sockets, RAM generation, board chipset, power draw, physical dimensions, cooling clearance, form factor, and upgrade data. That makes each recommendation inspectable and repeatable.
Before a build can be recommended, hard rules remove invalid combinations. The engine checks CPU socket, RAM type, Wi-Fi needs, PSU headroom, motherboard form factor, GPU length, cooler clearance, and PSU form factor. It only ranks a build after those checks pass.
The remaining compatible builds are scored with deterministic rules. Gaming profiles weight GPU capability and requested resolution; streaming, engineering, and creator profiles also reward their relevant CPU, graphics-memory, and acceleration traits. The result can compare value, performance, upgrade-friendly, and budget-saving approaches without letting a model invent hardware choices.
BuildAdvisor AI groups static-estimate budgets into five realistic ranges: $500–$700, $700–$900, $900–$1,300, $1,300–$1,700, and $1,700+. If a request does not fit—such as 4K gaming under $1,000—the result explains the mismatch and suggests a more realistic resolution, a higher budget, or carefully checked used parts.
Upgrade guidance is a transparent 10-point score, not a promise about future products. It considers newer AM5 and DDR5 platforms, motherboard upgrade capacity, power-supply room, and memory capacity. The result then explains which upgrade would make the most sense later, such as moving to 32GB of RAM or choosing a stronger GPU.
This is a prototype with local sample data, so prices are estimates rather than live store prices. It does not check real stock, exact benchmarks, used parts, regional availability, or the precise compatibility of a part a user already owns. It is designed to teach the decision process, not replace final manufacturer checks.
A future version could add model-specific owned-part checks, more detailed case clearance rules, guided game-setting expectations, saved builds, and optional current-market data with clear sources. The same rule-based compatibility layer would remain in place so any future recommendation stays explainable.
Design principle
BuildAdvisor AI answers with visible inputs, hard compatibility checks, transparent scoring priorities, budget reality checks, and a practical upgrade path. That makes it a small but real engineering system—not a black box that happens to produce a parts list.