{
  "id": 12026502,
  "title": "I Fine-Tuned an Open Model to Be My Friend's Patient Interview Coach",
  "url": "https://urgent.news/2026/10/04/i-fine-tuned-an-open-model-to-be-my-friends-patient-interview-coach",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-04T23:30:36.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/ishavishwakarma29/i-fine-tuned-an-open-model-to-be-my-friends-patient-interview-coach-3d3p"
  },
  "original_language": "en",
  "account": "InterviewBuddy is an open-weight model that has been fine-tuned specifically to function as a patient mock interviewer for a friend named Rose who is preparing for software engineering interviews. Rose has the ability to understand technical concepts, but often struggles to remember important details like the outcome of her stories during interviews. This mock interviewer works like a good practice partner, asking one question at a time and grading every answer using the same 4-part format: What worked, To improve, Stronger version, and Next question. If the user struggles with certain aspects of the interview, such as being nervous or rambling, the model provides helpful hints and cues instead of simply giving away the answer. The model also remembers past sessions and uses that information to target specific weak spots in future sessions. The model was built using an open-weight model that was fine-tuned using a LoRA adapter on top of the Qwen3-8B model. The training process involved creating 105 multi-turn mock-interview conversations across four categories, resulting in 153 graded feedback turns. The fine-tuning process took around 150K training tokens and cost a small fraction of the training credits. The end result is a personalized and tailored model designed to help Rose improve her interview skills, with the added benefit of being able to add her real sticking points to the dataset and retrain the model as she progresses.",
  "summary": "This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend What I Built InterviewBuddy is a patient mock interviewer for my friend Rose , who is preparing for software engineering interviews. Rose knows her stuff, but in interviews she freezes up and forgets to mention the result of her stories. Practicing with a real person is awkward and hard to schedule, and generic…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}