AI Resume Screening Showed 92% Match, But the Decision Was Already Made
Story of a senior candidate who was put through seven interview rounds over two weeks straight, even though the vacancy was closed a week earlier. AI screening showed 92% fit for the position. But it turned out: a neighboring department manager brought their own person, who was already getting access. The candidate served as insurance — if the preferred candidate bails, there's a backup. A week later came the rejection letter. The story shows: AI in recruitment often masks hiring problems rather than solving them.
AI-processed from Habr AI; edited by Hamidun News
A story of one interview reveals a problem in AI recruitment that gets discussed less than bias.
Candidate Misha
Misha is a senior with a solid resume. His resume passed AI screening: "92% match for the position" — the company was proud of implementing smart ATS. Then came: screening, technical interview, system design, live coding, cultural interview, conversation with the director, the finale. Two weeks of interviews. He solved a puzzle about elevators, drew diagrams, and told five times why he left his previous job.
What Misha didn't know
The vacancy was closed a week before the final round.
Vadim, the head of an adjacent department, brought his own person: "I've worked with him for three years, I vouch for him." That person just had tea with the director — that was the entire interview. He was already getting access while Misha was rehearsing his answer about conflict situations.
They kept putting Misha through more rounds. Why?
1. Reporting: the funnel, stages, scores all need to be in the system 2. Insurance: what if the internal candidate falls through, gets a competing offer, anything — Misha is waiting on the bench
They didn't tell him.
The role of AI
AI screening was perfect: 92% match. But it didn't solve the hiring problem — it masked it. The system says: "Look, this is an objective selection," when in reality the decision had already been made behind closed doors.
Vadim's internal hire was from a completely different tech stack. He ramped up in a couple of months, works fine. His experience had nothing to do with the position.
What this means
AI in recruitment often becomes cosmetics. Instead of fairly evaluating candidates, the system creates an illusion of objectivity, while real decisions are made through closed channels: referrals, internal candidates, connections. AI takes the blame for rejections, but doesn't participate in the choice.
Misha's story isn't about algorithm bias. It's about how AI often becomes a facade that companies need more for reporting than for fairness.
- Oct 28, 2026AdaPRA lifts Qwen 80B by 24.5 points on data science leaderboards
- Oct 28, 2026Anthropic reveals how it builds Claude with Claude: the brain separate from the hands
- Oct 28, 2026Anthropic Turns Claude Code Into a Platform: Claude Mods Flip the Agent Harness
- Oct 28, 2026Security engineer builds an MCP server in 7 days that spooks an AI agent with vulnerabilities
Key Questions Answered
What are the key details on Candidate Misha?
Misha is a senior with a solid resume. His resume passed AI screening: "92% match for the position" — the company was proud of implementing smart ATS. Then came: screening, technical interview, system design, live coding, cultural interview, conversation with the director, the finale. Two weeks of interviews. He solved a puzzle about elevators, drew diagrams, and told five times why he left his previous job.
What Misha didn't know?
The vacancy was closed a week before the final round. Vadim, the head of an adjacent department, brought his own person: "I've worked with him for three years, I vouch for him." That person just had tea with the director — that was the entire interview. He was already getting access while Misha was rehearsing his answer about conflict situations. They kept putting Misha through more rounds. Why? 1. Reporting: the funnel, stages, scores all need to be in the system 2. Insurance: what if the internal candidate falls through, gets a competing offer, anything — Misha is waiting
What are the key details on The role of AI?
AI screening was perfect: 92% match. But it didn't solve the hiring problem — it masked it. The system says: "Look, this is an objective selection," when in reality the decision had already been made behind closed doors. Vadim's internal hire was from a completely different tech stack. He ramped up in a couple of months, works fine. His experience had nothing to do with the position.
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.