Voice AI Interview — Fine-Tuned DistilGPT2
A causal language model fine-tuned on structured mock interview prompt-completion pairs for the Voice AI Interview project.
Model Details
- Base Model: distilgpt2
- Task: Text generation (AI interviewer responses)
- Fine-tuned on: Custom structured interview Q&A dataset
- Hardware: Apple Silicon MPS GPU via PyTorch
- License: Apache 2.0
Usage
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("rajatverma777/voice-ai-interview-gpt2")
model = AutoModelForCausalLM.from_pretrained("rajatverma777/voice-ai-interview-gpt2")
prompt = "Interviewer: Explain what is a binary search tree.\nCandidate:"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=150)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
Project
Part of the Voice AI Interview Assistant — a full-stack AI-powered mock interview platform.