Engineer with a proven track record of solving complex problems at scale — from raw data ingestion to fully deployed machine learning systems.
Expert in data science, machine learning, and full-stack development, with expertise in stats, mathematics, programming and computer systems.

I specialize in:

• Designing end-to-end ML pipelines for time series, tabular, and spatial data.
• Performing signal processing and feature engineering for predictive models.
• Building scalable data workflows that integrate cleanly with production systems.
• Bridging the gap between research prototypes and production-grade deployments.

Fluent in Python, SQL, and distributed data processing, with hands-on experience in Tensorflow, Spark, and Databricks.
I move fast, write clean code, and approach problems with first-principles thinking — delivering solutions that are practical, efficient, and production-ready.

📍 San Francisco, CA
📧 naman.bansal@wiliot.com
🔗 LinkedIn | GitHub

░▒▓█▓▒░░▒▓█▓▒░░▒▓█▓▒░▒▓█▓▒░▒▓█▓▒░      ░▒▓█▓▒░░▒▓██████▓▒░▒▓████████▓▒░ 
░▒▓█▓▒░░▒▓█▓▒░░▒▓█▓▒░▒▓█▓▒░▒▓█▓▒░      ░▒▓█▓▒░▒▓█▓▒░░▒▓█▓▒░ ░▒▓█▓▒░     
░▒▓█▓▒░░▒▓█▓▒░░▒▓█▓▒░▒▓█▓▒░▒▓█▓▒░      ░▒▓█▓▒░▒▓█▓▒░░▒▓█▓▒░ ░▒▓█▓▒░     
░▒▓█▓▒░░▒▓█▓▒░░▒▓█▓▒░▒▓█▓▒░▒▓█▓▒░      ░▒▓█▓▒░▒▓█▓▒░░▒▓█▓▒░ ░▒▓█▓▒░     
░▒▓█▓▒░░▒▓█▓▒░░▒▓█▓▒░▒▓█▓▒░▒▓█▓▒░      ░▒▓█▓▒░▒▓█▓▒░░▒▓█▓▒░ ░▒▓█▓▒░     
░▒▓█▓▒░░▒▓█▓▒░░▒▓█▓▒░▒▓█▓▒░▒▓█▓▒░      ░▒▓█▓▒░▒▓█▓▒░░▒▓█▓▒░ ░▒▓█▓▒░     
 ░▒▓█████████████▓▒░░▒▓█▓▒░▒▓████████▓▒░▒▓█▓▒░░▒▓██████▓▒░  ░▒▓█▓▒░     

Languages & Tools:

• Python (Pandas, NumPy, SciPy, PySpark, scikit-learn, PyTorch)
• SQL (BigQuery, Databricks SQL, Postgres)
• JavaScript (Node.js, Vanilla JS, Django + JS integrations)

Machine Learning / Data Science:

• Time Series Event Detection
• Gaussian Process Regression for Localization
• Cosine Similarity / Vectorization for asset-bridge mapping
• Feature Engineering on RF packet streams
• Model Deployment (Databricks MLflow, REST APIs)

Data Engineering:

• ETL pipelines in Databricks / Spark
• Large-scale data querying & optimization
• Real-time packet ingestion & enrichment

Web & Tools:

• Django, TailwindCSS, HTML/CSS/JS
• Git, CI/CD workflows, API integrations

Cloud & Infra:

• AWS (S3, Lambda, EC2)
• GCP (BigQuery, Cloud Functions)

• Text2SQL Platform for Conversing with Supply Chain Data

  • Built a production-level conversational AI platform enabling users to query real-time supply chain data in natural language.
  • Implemented multi-tenant organization access, role-based security, and a polished, responsive UI.
  • Integrated LLM frameworks including LangChain and LangGraph for query orchestration and SQL generation.
  • Deployed on scalable cloud infrastructure with real-time response and error-tolerant parsing.

• Temporal Event Detection Algorithms for IoT Devices

  • Designed and deployed a production-grade ML model for detecting temporal events in IoT asset data.
  • Achieved very high recall and high precision, with an almost perfect F1 score in live environments.
  • Engineered robust feature pipelines to handle noisy packet streams and large-scale time series processing.

• Retail IoT Localization Model

  • Designed top-k bridge RSSI fusion pipeline to improve asset location accuracy without dense bridge infrastructure.
  • Integrated pseudo-distance trilateration and signal-based features, outperforming heuristic baselines.

• Quizzx.com

  • Designed and deployed an interactive quiz platform with dynamic question generation.
  • Scalable backend with low-latency API responses and optimized data storage.

• COVID Vaccine Alerting System

  • Created a nationwide alerting platform to notify users of vaccine slot availability in real time.
  • Integrated multiple public APIs, high-throughput schedulers, and email push notifications.