About Me
I’m a data scientist and ML engineer based in Boston. I moved from chemical engineering to industrial engineering and then to machine learning. What ties those fields together is building systems that turn messy, real-world data into decisions people can trust.
I’m currently looking for Data Scientist, ML Engineer, and MLOps roles.
Contact
- Email: akshaypatelnitb6@gmail.com
- LinkedIn: linkedin.com/in/aksh-ay06
- GitHub: github.com/aksh-ay06
- Location: Boston, MA
Experience
Researcher / Data Analyst, Grant Fund Implementation — Data Driven WV
July 2026 – Present · Morgantown, WV
- Own the end-to-end data science lifecycle across two state grant programs: framing the analytical questions, preparing data, building and validating models, and delivering decision-support tools.
- Apply clustering and classification to identify funding priorities and target populations, and turn the results into recommendations for state agencies, community organizations, and funders.
- Build Tableau dashboards and ArcGIS geospatial tools that show funding allocation, program impact, and statewide broadband coverage gaps to non-technical stakeholders.
Cybersecurity Analyst — Data Driven WV
August 2025 – December 2025 · Morgantown, WV
- Led a team of 4 building an NLP decision-support system for NIST 800-53 compliance that automated the security audit workflow and cut manual processing time by 30%.
- Built a hybrid semantic and rule-based pipeline: MiniLM-L6-v2 embeddings with cosine similarity, plus deterministic validation in PostgreSQL to make control mapping more precise.
- Designed an on-prem backend with FastAPI and Docker, and a React dashboard that shows confidence scores and past security gaps.
Graduate Research Assistant — West Virginia University, IMSE Department
January 2024 – December 2025 · Morgantown, WV
- Led a team of 3 conducting data-driven energy audits for 9 commercial and industrial facilities.
- Built baseline consumption and retrofit simulation models using multivariate regression combined with engineering economics (ROI, payback period).
- Identified 1M+ in USDA REAP grant funding for facility upgrades.
- Turned the findings into executive dashboards and grant-compliant technical reports.
Education
- M.S. Industrial Engineering — West Virginia University · Jan 2024 – May 2026 Coursework projects: machine learning, quality and reliability engineering (multivariate SPC with PCA and Hotelling’s T²), manufacturing defect detection.
- B.Tech. Chemical Engineering — National Institute of Technology (NIT) Bhopal · Aug 2019 – May 2023
Technical Skills
| Area | Tools |
|---|---|
| Languages | Python, SQL, C++, JavaScript |
| ML / Stats | scikit-learn, PyTorch, XGBoost, LightGBM, statsmodels, samplics, SHAP |
| Causal Inference | DoWhy, EconML (DML, AIPW), propensity score methods, sensitivity analysis |
| LLM / GenAI | LangChain, LangGraph, Ollama, RAG, OpenSearch hybrid search, Langfuse, MCP |
| MLOps | MLflow, Airflow, Prefect, Kestra, Docker, FastAPI, Prometheus, drift detection |
| Data Engineering | PySpark, dbt, DuckDB, BigQuery, PostgreSQL, MongoDB, Redis |
| Cloud & Infra | GCP (Cloud Run, BigQuery, GCS), AWS, Terraform |
Selected Projects
See Projects for full write-ups. Highlights:
- arXiv RAG: a 10-service RAG system with hybrid retrieval and a local LLM.
- PMMDS: ML monitoring with PSI drift detection and automated champion/challenger retraining.
- Grupo Bimbo Forecasting: LightGBM on 74M+ transactions, with quantile prediction intervals.
- BRFSS Pollution & Cancer: survey-weighted inference and ML on 433K respondents.
Currently Reading
- Build a Large Language Model (From Scratch), Sebastian Raschka (implementations on GitHub)
- Designing Data-Intensive Applications, Martin Kleppmann
Outside Work
Bowling, pool, working out, cooking, and reading about whatever has caught my interest lately (currently wine).