I am an AI/ML Engineer and Data Scientist with a Master of Data Science from Monash University and 2+ years of enterprise experience at Cognizant[cite: 1]. I specialize in building, fine-tuning, and deploying agentic LLM workflows, production-grade RAG systems, and robust ETL pipelines[cite: 1].
- π οΈ Core Focus: Agentic AI, HyDE RAG, fine-tuning (SFT), and high-throughput LLM orchestration[cite: 1].
- π Achievements: Global #9 on Kaggle Leaderboard for predictive classification & Databricks Certified in dbt Analytics Engineering[cite: 1].
- π Location: Melbourne, Australia β‘οΈ Relocating to Bengaluru, India[cite: 1].
- π¬ Reach Me: ashwin.g2099@yahoo.com[cite: 1]
AI / ML & LLM Engineering
Backend & Data Engineering
Testing, Quality Engineering & Evaluation
Cloud, MLOps & Observability
π€ Sift β Production LLM Agent System
- Built a 5-agent orchestration engine using LangGraph, HyDE RAG, and pgvector[cite: 1].
- Fine-tuned
XLM-RoBERTa-baseas an intent pre-filter, cutting token consumption by 40%[cite: 1]. - Integrated LangFuse, LangSmith, and Prometheus for full production observability[cite: 1].
πΎ Kavach AI β Agentic Crop Insurance System
- Deployed a 9-agent LangGraph system backed by LLaMA 3.3 70B via Groq[cite: 1].
- Built an automated WhatsApp Business API chatbot delivering sub-30s insurance assessments based on live weather data[cite: 1].
πͺΈ ReefPulse AI β Real-Time Reef Health Monitor
- Engineered a 3-agent CrewAI system processing 372 live API sources at 60s intervals[cite: 1].
- Deployed a ML classifier achieving 0.97 Accuracy and 0.99 ROC-AUC with 7-day forecasting[cite: 1].
π AI-GitReviewer β Code Quality Gate
- Integrated Python AST analysis and HuggingFace CodeBERT to build a 0.90 precision quality gate with GitHub Actions CI/CD[cite: 1].