文件列表:

LICENSE (1070, 2023-06-06)
Submission (0, 2023-06-06)
Submission\testfile.py (575, 2023-06-06)
config.yaml (94, 2023-06-06)
context.txt (1565, 2023-06-06)
conversation.txt (2431, 2023-06-06)
interviewgpt.py (13994, 2023-06-06)
requirements.txt (320, 2023-06-06)
serviceAccountKey.json (2394, 2023-06-06)

# InterviewGPT AI leetcode interviewer that assesses tech applicants. Built on Langchain and OpenAI APIs. Recruiter-focused and tracks progress and submits analysis to email.  # Final features: **Interview content** - AI interviewer: provides focused and structured interviews, asking one question at a time and ensuring comprehension by asking follow-up questions before moving to the next question - Two question types: code-based and concept questions - AI interacts beyond simply asking the question; it follows up and asks you to elaborate depending on your answer. - CLI with context: user-friendly experience with relevant information and instructions - "Please enter a valid file path" - "You can type in the chat for this question" - Records transcript with code: - enables detailed review of candidate's coding abilities and problem-solving techniques - gives insight into what the candidate codes like - Variable interview length: adjusts based on context, usually around 45 minutes. - I only have access to GPT 3.5 API so outputs will be greatly improved when I get the new API **Backend** - Key validation on Firestore: enhances security, only registered candidates can access - Generate new keys - reference candidate emails **Candidate Results** - MIM-formatted email that prettifies the outputs so recruiters can read them easier - Email notification: prompt feedback, improves candidate communication - For the email text, I used langchain to summarize the transcript and provide scores, stats, and give verdict on candidate - timestamps to measure speed; you can see how long the candidate took for each question, possible flagging if they answer too fast **Prompting** - cut down Ack's original prompt to save tokens- AI interviewer: provides focused and structured interviews, asking one question at a time and ensuring comprehension by asking follow-up questions before moving to the next question

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：