Building intelligent systems that bridge research and production. 4 years crafting ML pipelines, LLM-powered applications, and scalable AI solutions that solve real-world problems.
I'm Baskar Robert, an AI Engineer with 4 years of hands-on experience designing and deploying machine learning systems across diverse domains.
My work spans the full AI lifecycle — from data engineering and model training to production deployment and monitoring. I specialize in building robust LLM applications, RAG pipelines, and intelligent automation systems.
I'm passionate about transforming complex AI research into practical, scalable products that deliver measurable business value. Whether it's fine-tuning foundation models or architecting multi-agent systems, I bring both depth and pragmatism to every challenge.
📍 Based in India · 🌐 www.baskarrobert.com
LLMs, RAG Systems, ML Engineering & AI Product Development
Professional AI/ML Engineering
Ship AI that actually works in production
Leading end-to-end development of LLM-powered applications including RAG systems and multi-agent pipelines. Architecting production-grade ML infrastructure serving real-time inference. Collaborating with cross-functional teams to deliver AI-driven product features. Mentoring junior engineers and establishing best practices for AI development.
Built and deployed machine learning models for classification, NLP, and recommendation systems. Developed data pipelines for model training and evaluation. Integrated ML models into REST APIs consumed by frontend teams. Conducted model performance monitoring and retraining workflows.
Performed exploratory data analysis and feature engineering for client datasets. Built predictive models for business use cases including churn prediction and demand forecasting. Created dashboards and visualizations to communicate insights to stakeholders.
A production-ready Retrieval-Augmented Generation system for enterprise knowledge bases. Supports multi-document ingestion, semantic chunking, and hybrid search with sub-second query latency across 100K+ document corpora.
End-to-end fine-tuning pipeline using QLoRA to adapt open-source LLMs for domain-specific tasks. Achieved 40% better task accuracy vs. zero-shot baselines, with automated evaluation and model versioning.
AI-powered document processing system that extracts structured data from unstructured PDFs and images using a combination of OCR, NLP, and LLM-based parsing. Processes 1000+ documents/hour with 96% extraction accuracy.
Customer churn prediction system using ensemble methods and SHAP-based explainability. Deployed as a real-time scoring API integrated with CRM systems, reducing churn by 18% through targeted interventions.
An autonomous multi-agent framework where specialized agents collaborate on complex research and analysis tasks. Features tool use, web search, code execution, and structured output generation.
Whether you have an AI project in mind, want to discuss collaboration, or just want to talk about the latest in LLMs — my inbox is always open.
hello@baskarrobert.com