Available to hire
Hi, I’m Atharva Gurav, a data-driven ML engineer and software developer with hands-on experience building scalable data pipelines, AI-powered automation, and production-ready inference services across academia and industry. I enjoy turning raw data into actionable insights, delivering measurable improvements in accuracy, latency, and reliability, and collaborating with cross-functional teams to ship impact-driven solutions.
When I’m not coding, I love learning new AI techniques, contributing to open-source projects, and mentoring teammates to translate complex data into clear business value.
Skills
Experience Level
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Work Experience
Data Engineer at IU Dining, Indiana University
August 1, 2023 - May 1, 2025Automated CRM workflow routing by matching 10K+ daily emails to opportunities; used vector similarity, hybrid ranking, and LLM disambiguation to personalize outputs, reaching 92% accuracy and cutting manual routing 70%. Built LangChain agentic orchestration for support diagnostics; executed plan–act–observe workflows with state tracking, tool-driven validations, and feedback loops, reducing incident response 60% with 99.5% integration reliability. Implemented real-time data-quality anomaly detection using 25+ validation rules and drift flagging; improved operational reporting.
Research Assistant at Indiana University Bloomington
January 1, 2025 - June 1, 2025Designed iterative ML training and evaluation pipelines using MLflow; applied ablation-driven feedback loops to improve mean IoU from 0.78 to 0.90 for production-ready model selection. Created reproducible data preprocessing workflows (Three.js labeling, PLY conversion) to standardize inputs; accelerated experiment cycles and improved evaluation consistency for downstream services. Deployed inference serving and benchmarked latency and memory across 10K+ point clouds; validated model behavior against operational constraints for reliable integration.
Machine Learning Engineer at Hyphenova
June 1, 2024 - November 1, 2024Developed multimodal classification for reasoning over BERT text embeddings and OpenL3 audio features; improved F1 from 0.68 to 0.82 (+20.6%) with 48ms p95 inference for production workflows. Performed large-scale hyperparameter tuning via grid search across 200+ configurations; refined model interactions and increased macro-F1 from 0.71 to 0.84. Built training and evaluation data ingestion for 100K+ posts with rate limiting and preprocessing; reduced manual labeling 40% and maintained reproducibility with Docker and MLflow.
Software Developer at Onefusion AI
March 1, 2022 - June 1, 2022Implemented TensorFlow LSTM time-series forecasting; automated feature engineering and hyperparameter optimization to deliver 20% projected cost reduction for utility stakeholders. Delivered client-facing ML workflows by collaborating on deployment and testing; documented assumptions and outcomes to support ongoing optimization of automated processes.
Education
MS in Computer Science at Indiana University Bloomington
August 1, 2023 - May 1, 2025BTech in Computer Science at Maharashtra Institute of Technology
July 1, 2019 - June 1, 2023Qualifications
Industry Experience
Software & Internet, Education
Skills
Experience Level
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Expert
Hire a Data Scientist
We have the best data scientist experts on Twine. Hire a data scientist in Bloomington today.