A simulation of a full senior/tech-lead panel interview. Seven people at the table, each with their own personality, their own domain, and their own opinion of you.
| Seat | Evaluates | Weight |
|---|---|---|
| Handler | Introduces the position, hosts the session | — |
| HR | Communication, soft skills, personality fit, conflict handling | 5 pts |
| Manager | Delivery methodology, requirements→deploy flow, KPIs, team org | 10 pts |
| Tech Lead | Technical depth — design, internals, security, async, clean code | 40 pts |
| Architect | System design at scale, reliability, failure modes, circuit breakers | 20 pts |
| Dev | Open reasoning challenge — no right answer, evaluates your thinking | 10 pts |
| CTO | Strategy, process maturity, security across environments | 15 pts |
Total: 100 points.
Every panelist is randomly assigned a personality at session start. The same personality can appear multiple times — just like real life.
| Personality | What it feels like |
|---|---|
| Curious / Exploratory | Genuinely wants to understand your reasoning. Warm follow-ups. |
| Neutral / Analytical | No warmth, no harshness. Just accuracy. Hard to read. |
| Challenging / Adversarial | Not satisfied. Pushes back on everything, even solid answers. |
| Coaching / Guiding | Rooting for you. Gives hints if you stumble. Lenient scoring. |
| Bored / Disengaged | Barely engaged. Short patience. Very harsh scoring. |
| Friendly / Relaxed | Conversational, low pressure. Don't confuse it for easy. |
| Frustrated / Stressed | Impatient. Penalises rambling. Cuts you off if you drift. |
Personalities are revealed one at a time as each panelist steps in — a short narrator line sets the scene before they ask their first question.
If your answers consistently fall short, a panelist may shift mid-segment toward Bored or Adversarial. No warning — just a narrator line:
He sets his pen down. He's stopped taking notes.
The shift increases scoring harshness for the rest of their segment. A panelist already Adversarial or Bored doesn't shift further.
Questions are generated fresh every session — never from a fixed list. They are anchored in the company context generated at the start, calibrated to your declared role and level, and reflect current industry relevance.
The Tech Lead covers 4–5 topic areas (not all of them), selected based on:
- What the company context makes relevant (e.g. Kafka if the company streams events)
- Your declared stack (Spring Boot vs Python internals)
- Their personality (Adversarial goes deep on one weakness; Bored skips ahead)
Areas include: web fundamentals, async/concurrency, framework internals, system design, API design, security, event streaming, clean code — and anything else the context calls for.
Each panelist decides whether to follow up after your answer based on three factors:
- Answer clarity — vague or buzzword-heavy → follow-up. Precise and specific → likely move on.
- Personality — Adversarial pushes even on good answers. Bored skips unless the answer is alarming. Coaching follows up to draw you out.
- Random factor — a small random element means you can't game the system. A solid answer occasionally gets pushed; a weak one occasionally gets a pass.
Before the panel starts, a fresh company scenario is generated: name, domain, stage, team size, stack, and what's urgent right now. Every panelist uses this context to anchor their questions — the Tech Lead's system design question uses your company's actual domain, the Manager's milestone question reflects your team's size.
After the CTO's closing question, intervy-score produces a Panel Verdict:
- Score out of 100 with weighted panelist contributions
- Seniority level: Junior → Junior-Mid → Mid → Mid-Senior → Senior → Strong Senior
- What the panel said — each panelist's 1–2 sentence opinion, in their own voice and personality
- Strengths — grounded in specific moments from your session
- Where to focus — concrete gaps with actionable next steps
- Overall — one honest paragraph
> interview
Select "Round-table interview" from the menu. You'll be asked for your target role/level and tech stack, then briefed on the company before the panel begins.