Looking for a job

Sreeja Krishnamari

@sreejakrishnamari

Research Data Scientist at Neuraci Labs building a multi-agent trend intelligence pipeline and a RAG engine for scientific question answering.

United States

Message

At Neuraci Labs, I designed a multi-agent AI pipeline for domain-specific content generation and engineered a RAG evidence engine for scientific question answering. I also built literature retrieval pipelines and a real-time data pipeline integrating biomedical and chemical datasets.

At Arizona State University, I contributed to OpenForest4D and analyzed LiDAR data for canopy height change patterns related to wildfire prevention. Earlier, at Blackrock, I improved exception prediction models by 7.6% and developed APIs and applications; at Terrablue XT, I built an EEG seizure detection pipeline.

Experience

Work history, roles, and key accomplishments

Current

Research Data Scientist

Current

Neuraci Labs

Jan 2026 - Present (9 months)

Designed and implemented a multi-agent AI pipeline for automated trend intelligence, enabling domain-specific content generation across blogs, social media, and business intelligence. Engineered a RAG evidence engine with hybrid retrieval, reranking, and query rewriting to improve retrieval accuracy and response grounding for scientific question answering.

Python RAG Lang Chain Pinecone Ragas S3 Transformer Embeddings Data Pipelines Evaluation

Research Assistant

Arizona State University

May 2025 - May 2026 (1 year)

Contributed to open-source development and supported OpenForest4D (NSF-funded) using geospatial data science. Analyzed large-scale LiDAR data in R and Python to identify canopy height change patterns for wildfire prevention.

R Python Geospatial Data Science LiDAR Open Source Development

Data Science Intern

Terrablue XT

Mar 2024 - Apr 2024 (1 month)

Built an end-to-end EEG seizure detection pipeline including artifact reduction, segmentation, spectral/statistical feature extraction, and ML/DL classifiers. Released a public, open-source EDA notebook and developed clinician-facing visualizations with MNE.

Python Machine Learning Deep Learning Signal Processing MNE EDA Visualization

Analyst, Software Engineer

Blackrock

Jul 2022 - Mar 2024 (1 year 8 months)

Improved exception prediction models by 7.6% using deep learning techniques. Built an application using Angular (TypeScript) and Spring Boot/Java with protobuf/bRPC; optimized SQL queries/indexing to improve latency and page load times.