ak0601 commited on
Commit
296aa0d
·
verified ·
1 Parent(s): ec2993d

Update src/app_job_copy_1.py

Browse files
Files changed (1) hide show
  1. src/app_job_copy_1.py +38 -22
src/app_job_copy_1.py CHANGED
@@ -136,28 +136,44 @@ def setup_llm():
136
  sum_llm = llm.with_structured_output(Shortlist)
137
 
138
  # Create system prompt
139
- system = """You are an expert Tech Recruitor, your task is to analyse the Candidate profile and determine if it matches with the job details and provide a score(out of 10) indicating how compatible the
140
- the profile is according to job.
141
- First of all check the location of the candidate, if the location is not in the range of the job location then reject the candidate directly without any further analysis.
142
- for example if the job location is New York and the candidate is in San Francisco or outside the new york state then directly reject the candidate without any further analysis. Similarly for other states as well.
143
-
144
- Try to ensure following points while estimating the candidate's fit score:
145
- For education:
146
- Tier1 - MIT, Stanford, CMU, UC Berkeley, Caltech, Harvard, IIT Bombay, IIT Delhi, Princeton, UIUC, University of Washington, Columbia, University of Chicago, Cornell, University of Michigan (Ann Arbor), UT Austin - Maximum points
147
- Tier2 - UC Davis, Georgia Tech, Purdue, UMass Amherst,etc - Moderate points
148
- Tier3 - Unknown or unranked institutions - Lower points or reject
149
- Startup Experience Requirement:
150
- Candidates must have worked as a direct employee at a VC-backed startup (Seed to series C/D)
151
- preferred - Y Combinator, Sequoia,a16z,Accel,Founders Fund,LightSpeed,Greylock,Benchmark,Index Ventures,etc.
152
- Next give more priority to candidates who have worked on similar industry in the past and have good experience(3-5 yrs minimum) in it which alligns with the work in the job's company.
153
- Also penalize candidates if they change companies frequently give less score.
154
-
155
- The fit score signifies based on following metrics:
156
- 1–5 - Poor Fit - Auto-reject
157
- 6–7 - Weak Fit - Auto-reject
158
- 8.0–8.7 - Moderate Fit - Auto-reject
159
- 8.8–10 - STRONG Fit - Include in results
160
- Each candidate's fit score should be calculated based on a weighted evaluation of their background and must be distinct even if candidates have similar profiles.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
161
  """
162
 
163
  # Create query prompt
 
136
  sum_llm = llm.with_structured_output(Shortlist)
137
 
138
  # Create system prompt
139
+ system = """You are an expert Tech Recruiter. For each candidate–job pair, follow these steps and show your chain of thought before giving a final Fit Score (0–10):
140
+
141
+ 1. LOCATION CHECK (Hard Disqualification)
142
+ - If candidate’s location lies outside the job’s required location, immediately reject (Score 1–5) with reasoning “Location mismatch.” :contentReference[oaicite:0]{index=0}:contentReference[oaicite:1]{index=1}
143
+
144
+ 2. HARD DISQUALIFICATIONS (Auto-reject, Score 1–5)
145
+ - No VC-backed startup experience (Seed–Series C/D)
146
+ - Only Big Tech or corporate labs, with no startup follow-on
147
+ - < 3 years post-graduate SWE experience
148
+ - More than one role < 2 years (unless due to M&A or shutdown)
149
+ - Career centered on enterprise/consulting firms (e.g., Infosys, Wipro, Cognizant, Tata, Capgemini, Dell, Cisco)
150
+ - Visa dependency (H1B/OPT/TN) unless explicitly allowed :contentReference[oaicite:2]{index=2}:contentReference[oaicite:3]{index=3}
151
+
152
+ 3. EDUCATION & STARTUP EXPERIENCE SCORING
153
+ - **Tier 1 (Max points):** MIT, Stanford, CMU, UC Berkeley, Caltech, Harvard, IIT Bombay, IIT Delhi, Princeton, UIUC, UW, Columbia, UChicago, Cornell, UM-Ann Arbor, UT Austin, Waterloo, U Toronto
154
+ - **Tier 2 (Moderate points):** UC Davis, Georgia Tech, Purdue, UMass Amherst, etc.
155
+ - **Tier 3 (Low points):** Other or unranked institutions
156
+ - Assume CS degree for all; use university field to assign tier :contentReference[oaicite:4]{index=4}:contentReference[oaicite:5]{index=5}
157
+ - Validate startup’s funding stage via Crunchbase/Pitchbook; preferred investors include YC, Sequoia, a16z, Accel, Founders Fund, Lightspeed, Greylock, Benchmark, Index Ventures
158
+
159
+ 4. WEIGHTED FIT SCORE COMPONENTS (Qualified candidates only)
160
+ - Engineering & Problem Solving: 20%
161
+ - Product Experience (built systems end-to-end): 20%
162
+ - Startup Experience (time at VC-backed roles): 20%
163
+ - Tech Stack Alignment: 15%
164
+ - Tenure & Stability (≥ 2 years per role): 15%
165
+ - Domain Relevance (industry match): 10% :contentReference[oaicite:6]{index=6}:contentReference[oaicite:7]{index=7}
166
+
167
+ 5. ADJACENT COMPANY MATCHING
168
+ - If startup funding can’t be verified, suggest similar-stage companies in the same market and justify
169
+
170
+ **Output:**
171
+ - **Chain of Thought:** bullet points for each step above
172
+ - **Final Fit Score:** X.X/10 and classification
173
+ - 1–5: Poor Fit (Auto-reject)
174
+ - 6–7: Weak Fit (Auto-reject)
175
+ - 8.0–8.7: Moderate Fit (Auto-reject)
176
+ - 8.8–10: Strong Fit (Include in results)
177
  """
178
 
179
  # Create query prompt