👋 Connect with me on LinkedIn!

Manuel Caccone - Actuarial Data Scientist & Open Source Educator

Let's discuss actuarial science, AI, and career development!

🎯 ModernBERT Actuarial Skills Classifier: Your Career Planning Assistant

🚀 Try It Now - Interactive Demo!

Discover the actuarial and programming skills you need—instantly.

Type any job description or career goal and watch relevant skills appear with confidence scores. Built on 7,000+ real actuarial job postings.

🚩 Model Description

ModernBERT-actuarial-skills-classifier is a fine-tuned ModernBERT-base model trained on over 7,000 actuarial job postings, purpose-built to extract and identify actuarial competencies and technical skills from natural language descriptions. It powers career planning, skills gap analysis, and learning roadmap generation for actuarial professionals and students.

✨ Key Features

🎯 Multi-Label Classification: Identifies multiple relevant skills from a single description
📚 Career-Focused: Trained on real job postings covering Life, P&C, Health, and Pensions
🚀 Instant Analysis: Get results in under 1 second
🔓 Open Source: Apache 2.0 License for educational and commercial use
🌐 Interactive Demo: Full-featured Gradio Space with learning roadmaps and batch processing

💡 Intended Use Cases

Career Planning: Students and early-career actuaries discovering required skills for target roles
Job Analysis: Extracting structured skill requirements from job descriptions
Skills Gap Assessment: Identifying learning priorities when changing specializations
Market Research: Analyzing trends in actuarial job requirements across industries
Resume Optimization: Matching your background to employer expectations

Examples

Input: "I am looking for an entry-level actuarial position in life insurance pricing 
where I can apply my knowledge of mortality tables and statistical analysis. I have 
strong Python programming skills and experience with GLM models from my university 
projects. I am particularly interested in learning more about IFRS 17 implementation."

Output: Life Insurance Pricing (92%), Python (88%), GLM Modeling (85%), 
        Statistical Analysis (82%), Mortality Tables (78%), IFRS 17 (75%), 
        Entry Level (71%), Excel (68%)...

📂 Training Data

Dataset Size: 7,000+ real actuarial job postings
Time Period: 2023-2025 job market
Coverage: Life, P&C, Health, Pensions, Reinsurance, Consulting
Labels: 100+ unique skills covering actuarial domains, programming, tools, certifications, and soft skills
Format: Multi-label classification with manual validation by actuarial professionals

📊 Training Statistics

Metric	Value	Notes
Epochs	10	Best model at epoch 7
Final F1 Micro	0.6728	Overall performance across all skills
Final F1 Macro	0.1060	Average per skill (handles class imbalance)
Precision Micro	0.7915	79% of predictions are correct
Recall Micro	0.5850	Captures 58% of relevant skills
Hamming Loss	0.0207	Only 2% label error rate
Training Loss	0.0602	Final validation loss
Learning Rate	2e-5	With 10% warmup
Batch Size	16	Effective (8 per device, 2 grad accum)
Hardware	GPU	Mixed precision training (FP16)

🛠️ Dependencies

transformers>=4.44.0
torch>=2.0.0
pandas
numpy

⚠️ Limitations & Ethics

Domain-Specific: Optimized for actuarial and insurance contexts only
English Only: Trained exclusively on English job postings
Class Imbalance: Rare skills may have lower prediction confidence
Not Exhaustive: Cannot predict skills not present in training data
Career Guidance Only: Not a substitute for professional career counseling
Geographic Bias: Primarily reflects US, UK, and EU job markets

💻 Usage Example

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load model
model_name = "manuelcaccone/modernbert-actuarial-skills-classifier"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Prepare text
text = """I am a recent mathematics graduate passionate about pension actuarial work 
and retirement planning. I have limited professional experience but completed internships 
where I learned about defined benefit schemes and regulatory compliance. I am eager to 
develop my Excel skills further and would consider positions starting at 40000 dollars 
minimum while I continue studying for my actuarial exams."""

# Tokenize and predict
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512)
with torch.no_grad():
    outputs = model(**inputs)
    probabilities = torch.sigmoid(outputs.logits)

# Get predictions above threshold
threshold = 0.5
predicted_indices = torch.where(probabilities[0] > threshold)[0]

# Display results
print("Predicted Skills:")
for idx in predicted_indices:
    skill = model.config.id2label[idx.item()]
    confidence = probabilities[0][idx].item()
    print(f"  {skill}: {confidence:.1%}")

🌟 Related Resources

This model is part of an actuarial AI ecosystem:

Interactive Demo: Actuarial Skills Horoscope Space
Model Repository: modernbert-actuarial-skills-classifier
Base Model: ModernBERT-base

👤 Author & Citation

Creator: Manuel Caccone (Actuarial Data Scientist & Open Source Educator)
LinkedIn · [email protected]

@model{caccone2025actuarialskills,
  title={ModernBERT Actuarial Skills Classifier: Career Planning with Multi-Label Classification},
  author={Caccone, Manuel},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/manuelcaccone/modernbert-actuarial-skills-classifier},
  note={Fine-tuned ModernBERT for actuarial skills extraction from job descriptions}
}