AI Roasts My Career

Analyzes user-provided career history, skills, and background through a prompt-engineered LLM pipeline designed to bypass corporate politeness filters and deliver candid, unfiltered feedback on career viability and trajectory. The system uses adversarial prompting techniques to force the model to critique rather than praise, generating assessments that highlight realistic gaps, market saturation, and misalignments between aspirations and qualifications without softening language or motivational framing.

Implements a minimal-friction input pipeline that accepts unstructured career information (job titles, years of experience, skills) and routes it directly to an LLM for analysis without requiring lengthy questionnaires or structured form completion. The system prioritizes speed over comprehensiveness, using a simple text submission interface that processes input through a single-pass LLM call and returns results within minutes rather than requiring multi-step assessment workflows.

Implements prompt-level constraints that force the LLM to adopt a critical, unfiltered voice by explicitly instructing the model to identify weaknesses, market saturation, and realistic limitations rather than defaulting to encouragement. The system uses negative framing instructions ('What's wrong with this career path?' rather than 'What are the strengths?') and explicitly disables politeness tokens to generate assessments that feel genuinely critical rather than diplomatically balanced.

Implements a completely free service model with no authentication, account creation, or payment processing, allowing users to submit career information and receive assessments without any friction or upsell mechanisms. The system is designed as a public utility rather than a lead-generation tool, with no email capture, no freemium tier, and no conversion funnel to paid services, making the assessment accessible to anyone with a web browser.

Processes career input through a single LLM inference call that generates a complete assessment in one pass, without follow-up questions, clarification loops, or iterative refinement. The system treats the assessment as a one-shot output rather than a conversation, meaning the LLM receives the user's input once and produces final feedback without the ability to ask for missing context, drill deeper into specific concerns, or adjust the analysis based on additional information.