Candidates are embedding hidden text and instructions in their resumes to game automated AI hiring platforms. This 'prompt hacking' tactic, reportedly found in up to 10% of applications by one firm, represents a new front in the cat-and-mouse game between applicants and the algorithms designed to filter them.

Contrary to expectations, job candidates found it easier to talk to an AI interviewer. The lower pressure of a non-human interaction allowed them to relax, be more open, and talk more freely about their experiences, leading to better outcomes.

Voice mode offers a more natural and effective way to practice for interviews than text-based AI. For best results, provide the AI with your resume and the job description for the role. This allows it to tailor questions, provide more relevant feedback, and simulate a real interview scenario.

To gauge an expert's (human or AI) true depth, go beyond recall-based questions. Pose a complex problem with multiple constraints, like a skeptical audience, high anxiety, and a tight deadline. A genuine expert will synthesize concepts and address all layers of the problem, whereas a novice will give generic advice.

To simulate interview coaching, feed your written answers to case study questions into an LLM. Prompt it to score you on a specific rubric (structured thinking, user focus, etc.), identify exact weak phrases, explain why, and suggest a better approach for structured, actionable feedback.

Beyond transcription, advanced AI tools can analyze an interviewer's live performance. They offer feedback on tonality, vocabulary, use of open vs. closed questions, and even body language, turning the AI into a powerful tool for improving human soft skills and communication.

Ditch standard FANG interview questions. Instead, ask candidates to describe a messy but valuable project they shipped. The best candidates will tell an authentic, automatic story with personal anecdotes. Their fluency and detail reveal true experience, whereas hesitation or generic answers expose a lack of depth.

Ineffective interviews try to catch candidates failing. A better approach models a collaborative rally: see how they handle challenging questions and if they can return the ball effectively. The goal is to simulate real-world problem-solving, not just grill them under pressure.

Upload interview transcripts and a job description into an AI tool. Program it to define the top criteria for the role and rate each candidate's transcript against them. This provides an objective analysis that counteracts personal affinity bias and reveals details missed during the live conversation.