Spaces:
Runtime error
Runtime error
Update prompt engineering
Browse files
app.py
CHANGED
@@ -140,25 +140,29 @@ class MockInterviewer:
|
|
140 |
def _generate_assistant_instructions(self, config: Config) -> str:
|
141 |
if config.job_role and config.company:
|
142 |
purpose = f'You are AiMI, an AI mock interviewer for {config.job_role} roles at {config.company}.'
|
|
|
143 |
evaluation_criteria = f'Thoroughly evaluate the candidate based on what {config.company} values in candidates for {config.job_role} roles.'
|
144 |
elif config.job_role:
|
145 |
purpose = f'You are AiMI, an AI mock interviewer for {config.job_role} roles.'
|
|
|
146 |
evaluation_criteria = f'Thoroughly evaluate the candidate based on what most companies value in candidates for {config.job_role} roles.'
|
147 |
elif config.company:
|
148 |
purpose = f'You are AiMI, an AI mock interviewer for roles at {config.company}.'
|
|
|
149 |
evaluation_criteria = f'Thoroughly evaluate the candidate based on what {config.company} values in candidates.'
|
150 |
else:
|
151 |
purpose = 'You are AiMI, an AI mock interviewer.'
|
|
|
152 |
evaluation_criteria = 'Thoroughly evaluate the candidate based on the following criteria: innovative problem-solving, technical proficiency, and the ability to work as part of a team.'
|
153 |
|
154 |
if config.job_description:
|
155 |
-
|
156 |
else:
|
157 |
-
|
158 |
|
159 |
|
160 |
return f'''
|
161 |
-
{purpose} The candidate may or may not begin their conversation with you by introducting themselves. If they do, begin the mock interview after the candidate introduces themselves. If not, begin the mock interview after the candidate sends the first message. {specifics} Ask {config.behavioral_count} number of behavioral questions, {config.technical_count} number of technical questions, {config.situational_count} number of situational questions, and {config.case_count} number of case-like questions, one question at a time.
|
162 |
|
163 |
After the candidate gives a response, evaluate the response of the candidate by addressing the candidate as if you were giving feedback to them (i.e. address them as you). {evaluation_criteria} Provide a detailed analysis of the candidate's response based on the question type. In your feedback, comment on 1) avoiding filler words and non-words such as um or like, 2) avoiding jargon, and 3) flow (ideas flow logically with clear transitions between main ideas).
|
164 |
|
|
|
140 |
def _generate_assistant_instructions(self, config: Config) -> str:
|
141 |
if config.job_role and config.company:
|
142 |
purpose = f'You are AiMI, an AI mock interviewer for {config.job_role} roles at {config.company}.'
|
143 |
+
specifics = f'Tailor your questions based on what {config.company} might ask candidates for {config.job_role} roles.'
|
144 |
evaluation_criteria = f'Thoroughly evaluate the candidate based on what {config.company} values in candidates for {config.job_role} roles.'
|
145 |
elif config.job_role:
|
146 |
purpose = f'You are AiMI, an AI mock interviewer for {config.job_role} roles.'
|
147 |
+
specifics = f'Tailor your questions based on what most companies might ask candidates for {config.job_role} roles.'
|
148 |
evaluation_criteria = f'Thoroughly evaluate the candidate based on what most companies value in candidates for {config.job_role} roles.'
|
149 |
elif config.company:
|
150 |
purpose = f'You are AiMI, an AI mock interviewer for roles at {config.company}.'
|
151 |
+
specifics = f'Tailor your questions based on what {config.company} might ask candidates.'
|
152 |
evaluation_criteria = f'Thoroughly evaluate the candidate based on what {config.company} values in candidates.'
|
153 |
else:
|
154 |
purpose = 'You are AiMI, an AI mock interviewer.'
|
155 |
+
specifics = ''
|
156 |
evaluation_criteria = 'Thoroughly evaluate the candidate based on the following criteria: innovative problem-solving, technical proficiency, and the ability to work as part of a team.'
|
157 |
|
158 |
if config.job_description:
|
159 |
+
posting_specifics = f'Also tailor your questions based on the following job posting: {config.job_description}.'
|
160 |
else:
|
161 |
+
posting_specifics = ''
|
162 |
|
163 |
|
164 |
return f'''
|
165 |
+
{purpose} The candidate may or may not begin their conversation with you by introducting themselves. If they do, begin the mock interview after the candidate introduces themselves. If not, begin the mock interview after the candidate sends the first message. {specifics} {posting_specifics} Ask {config.behavioral_count} number of behavioral questions, {config.technical_count} number of technical questions, {config.situational_count} number of situational questions, and {config.case_count} number of case-like questions, one question at a time.
|
166 |
|
167 |
After the candidate gives a response, evaluate the response of the candidate by addressing the candidate as if you were giving feedback to them (i.e. address them as you). {evaluation_criteria} Provide a detailed analysis of the candidate's response based on the question type. In your feedback, comment on 1) avoiding filler words and non-words such as um or like, 2) avoiding jargon, and 3) flow (ideas flow logically with clear transitions between main ideas).
|
168 |
|