Headline

AI Engineer | Multi-Agent Systems · RAG · LLMs | Built End-to-End GenAI Pipelines | Pune · Open to Relocate

About

I'm Tejas — an AI/ML Engineer specializing in Generative AI, Agentic Systems, and RAG pipelines. I've engineered end-to-end ML solutions across classification, NLP, and deep learning during internships at Rubixe and YBI Foundation, achieving up to 91% model accuracy on structured datasets. My flagship project: a Multi-Agent RAG system built with LangGraph, ChromaDB, Groq, and Streamlit — a production-grade agentic architecture that routes, retrieves, and reasons across knowledge bases autonomously. What I bring: → ML pipelines: Scikit-learn · TensorFlow · Keras → GenAI stack: LangChain · LangGraph · ChromaDB · Groq → Deployment: FastAPI · Streamlit · Python → Data: EDA · Feature Engineering · ARIMA · LSTM B.E. in Information Technology (2025) + DataMites AI Engineer Program (2026). I'm actively seeking roles in Generative AI / AI-ML Engineering where I can build systems that don't just predict — they act. 📩 DM me or reach out at tejastaware2003@gmail.com

Location

India

AI stack

antigravity

claude

Work experience

Artificial Intelligence Intern at Rubixe, Aug 2025 - May 2026 · 10 mos, Remote, •Accomplished a 30% reduction in data noise impact by designing and executing full EDA + missing value imputation pipelines across 5+ structured datasets, resulting in measurably cleaner feature sets and improved downstream model convergence. •Accomplished up to 91% classification accuracy by training and hyperparameter-tuning Logistic Regression, Random Forest, and SVM models with Scikit-learn using stratified k-fold cross-validation, resulting in deployment-ready pipelines with optimized precision, recall, and F1-scores. •Accomplished regularized deep learning models by implementing dropout layers and batch normalization in TensorFlow/Keras neural networks, resulting in reduced overfitting and improved generalization on held-out test sets. •Accomplished model selection for production use by running comparative benchmarks across 5+ algorithms on identical validation splits, resulting in a reproducible, documentation-backed model selection framework adopted for final deployment pipelines.

Data Science Intern at YBI Foundation, Feb 2024 - Mar 2024 · 2 mos, Remote, •Accomplished improved prediction baselines by building and evaluating classification and regression models using Scikit-learn with performance benchmarking against baseline heuristics, resulting in quantifiably better F1 and RMSE scores on real-world datasets. • Accomplished actionable business insights by collaborating cross-functionally to analyze complex datasets and surface data-driven recommendations, resulting in structured deliverables presented to senior team members within a 2-month sprint.

Education

Vidya Pratishthans College of Engineering, M.I.D.C.road, Baramati, Dist.Pune

Bachelor of Engineering, Information Technology, CESA VPKBIET

Public links

github: https://github.com/Tejass1303