I build applied AI systems that turn complex information and workflows into useful products—from multilingual model serving and document intelligence to developer tooling and forecasting.
- Growing an open-source practice across AI infrastructure, developer tooling, and applied ML
- AWS Certified · MS in Business Analytics (Data Science), UT Dallas
- Based in Dallas, focused on agentic systems, GenAI infrastructure, and developer tools
Analyzes repository issues, surfaces priority and complexity insights, and orchestrates automated fix and pull-request workflows.
Python · FastAPI · Streamlit · OpenAI · Composio
A SageMaker inference system for Spanish, French, and Russian support, with dynamic LoRA adapter loading, batching, and resource cleanup.
AWS SageMaker · LoRA · Python · S3 · LMI
Full-stack document intelligence combining semantic retrieval, OCR, layout understanding, and LLM-powered analysis.
Qdrant · BGE · Mixtral · LayoutLMv3 · OCR
- Intermittent Demand Forecasting — compared ARIMA, Croston, exponential smoothing, and Prophet for sparse spare-parts demand
- Rice Crop Prediction — combined machine learning and satellite data to identify cultivation areas and forecast crop yield
- Sentiment Analysis with Deep Learning — explored NLP techniques for classifying text sentiment
Open source is becoming a larger part of my engineering practice. I am interested in contributing across ecosystems where AI infrastructure, developer experience, model serving, and applied machine learning meet.
I care about contributions that are technically useful and maintainable: focused fixes, thoughtful tests, clear documentation, and tooling that makes complex systems easier to use.
| Area | Technologies |
|---|---|
| AI systems | Python, PyTorch, Hugging Face, LangChain, RAG, LoRA |
| Cloud & platform | AWS SageMaker, Bedrock, S3, Docker, FastAPI, Qdrant |
| Product engineering | TypeScript, Next.js, React, Streamlit |
| Data & modeling | pandas, scikit-learn, SQL, time-series forecasting, Plotly |
- Sujithra's Field Notes — An independent, source-backed journal about AI systems, software, data, infrastructure, and interaction design · Source
- Demystifying P-Values: A Guide for Non-Technical Stakeholders
- Understanding Type I and Type II Errors in Hypothesis Testing
- All About Data Preprocessing
Interested in applied AI, open source, or building intelligent products?
Let's connect on LinkedIn →