Building Intelligent AI Systems | Data Science β’ Automation β’ Machine Learning β’ MLOps β’ RAG β’ Orchestration
Iβm Aarjun Mahule, an aspiring Data Scientist & AI Engineer passionate about building production-grade AI solutions and end-to-end machine learning systems.
I specialize in designing intelligent pipelines that collect, process, analyze, and act on data autonomously β from predictive ML models to fully automated multi-agent workflows. My work blends AI, automation, orchestration, and infrastructure to deliver solutions that scale.
I aim to build a global career that merges deep tech engineering, freelancing, and entrepreneurship, and Iβm relentlessly focused on mastering the technologies that power modern AI systems.
- Python (primary), SQL, Shell scripting
- Scikit-learn, TensorFlow, PyTorch, Deep Learning, NLP, Feature Engineering, Model Evaluation, Hyperparameter Tuning
- Pandas, NumPy, Matplotlib, Seaborn, Data Cleaning, EDA, Data Visualization, API Data Pipelines
- Playwright (headless browser automation), n8n (workflow orchestration), FastAPI (backend APIs), REST API design, Task Queues (Celery), Redis
- Docker, GitHub Actions (CI/CD), Terraform, API Deployment, Experiment Tracking, Logging & Monitoring (Prometheus, Grafana, Sentry)
- PostgreSQL, SQLite, ChromaDB (Vector DB), Data Modeling, Indexing
- Retrieval-Augmented Generation (RAG), Embeddings, ChromaDB pipelines, Multi-agent AI architecture
- Git, GitHub, VS Code, Power BI, Tableau, Hugging Face, AWS, Notion API, Gmail API
- π§ AI Job Application Agent β A production-grade AI system that autonomously discovers, filters, applies, and tracks 1000+ job opportunities using multi-agent orchestration, Playwright automation, RAG pipelines, and intelligent scoring logic.
- π³ Credit Card Fraud Detection β Machine learning pipeline with SMOTE balancing, 284K+ transactions, and a 0.99 F1-score, built for high-performance anomaly detection.
- π¦ Wildlife Movement Prediction β GRU-based deep learning model that predicts bird migration patterns using spatio-temporal data from the Movebank API, featuring custom preprocessing and sequential modeling.
"Build relentlessly. Automate intelligently. Scale fearlessly."
When Iβm not coding, I explore AI trends, study automation systems, and brainstorm ideas for the next generation of tech products.
βοΈ If you find my work interesting, consider starring my repositories β it motivates me to build even more ambitious projects!