Pick a round. Sokrates asks one question at a time, transcribes your answer as you speak, and scores it out of 10 with what worked, what to change, and what you missed.
Included on the free plan · every session saved to your history
Coding, system design, ML / data science, behavioral, intro, HR, general, or your own topic. Set the difficulty and focus areas.
No preamble, no “great answer”. Technical rounds get fresh, standard-format problems; behavioral rounds ask about your own experience.
Your answer is transcribed live. When you are done, the feedback arrives: a score out of 10, then Worked, Improve, Missed, and a model answer.
An overall score with one-line call-outs on clarity, depth, and structure, plus strengths and what to tighten next time.
The same round types as interview mode, so the answer shape you practise is the one you get on the day.
Warm-up, then behavioral, then technical — the way a real loop goes.
“Tell me about yourself”, why this role, salary and logistics.
Situation, action, result, lesson — one story per question.
Fresh, self-contained problems at the difficulty you set.
Requirements, API, data model, components, scaling, failure modes.
Concepts, modelling choices, and ML system design.
Values, working style, why-us — answered as you, not a template.
Write your own topic and focus areas.
Every score out of 10 is anchored: 5 is “would not pass at this level”, 6–7 is borderline with real gaps, 8+ is a clear hire signal. A skipped or one-line answer scores 3 or less. The verdict quotes your actual words, the Improve line is a sentence you could say next time, and the model answer shows what a strong candidate would have said.
Practise in another language and the questions, transcription, and feedback all switch with it.
Practice is included on the free plan. Ten minutes of live transcription a day; Pro removes the cap.