Anish Maitra

Experienced Data Professional with a robust foundation in Full Stack Data Analytics, combining deep expertise in Data Engineering, Machine Learning, Cloud Infrastructure, and Generative AI. Proven ability to design and scale end-to-end data pipelines, manage cloud-based data platforms, and deploy AI/ML models for production-grade analytics.

Hands-on experience in time series forecasting, supervised learning, and deep learning with frameworks like TensorFlow and PyTorch. Skilled in data-centric AI, optimizing ML model performance through thoughtful data transformation, warehousing, and schema design. Recently contributed to Generative AI projects, including LangChain-based RAG systems, prompt engineering, and LLM-powered automation.

An adaptable and collaborative team player committed to solving real-world problems at the intersection of Data Engineering and AI, consistently driving business impact through actionable insights and smart automation.

1. Retail Demand Forecasting Using Automated ML (AutoML)

Role: Data Science Analyst

Industry: Retail

Duration: June 2022 – April 2023

Tech Stack: PyCaret, Prophet, LightGBM, Python, Tableau

Summary:

Addressed inventory planning inefficiencies for a large retail client using time series forecasting and AutoML techniques.

Highlights:

• Used PyCaret’s AutoML to identify optimal models across multiple store-product combinations.
• Evaluated and compared models including ARIMA, Prophet, and LightGBM; deployed top performers in production.
• Developed a dynamic Tableau dashboard for inventory managers, visualizing demand trends and forecast-based recommendations.
• Achieved a 20% reduction in stockouts, and a 30% improvement in demand forecast precision.

2. Building Datalake Platform for ML and BI

Role: Data Engineer

Industry: FMCG / Retail

Duration: April 2023 – April 2024

Tech Stack: PostgreSQL, Snowflake, SnapLogic, Apache Airflow, AWS Glue

Summary:

Engineered a modern Datalake architecture to support enterprise analytics and machine learning enablement.

Highlights:

• Designed and built end-to-end ETL pipelines from PostgreSQL to Snowflake using SnapLogic and AWS Glue.
• Orchestrated pipelines via Apache Airflow DAGs for monitoring and dependency management.
• Applied best practices in Data Warehouse Modeling using Star and Snowflake schemas, and created domain-specific data marts.
• Collaborated with Data Science teams to deliver ML-ready feature sets, leading to up to 18% improvement in model accuracy.
• Enabled seamless BI access using Tableau, and supported ML model pipelines directly from the Datalake layer.

3. Cloud Infrastructure Management and DevOps.

Role: DevOps Analyst

Industry: Cross-functional / Analytics Platform

Duration: April 2024 – September 2024

Tech Stack: AWS (EC2, ALB), Azure DMS, Power BI

Summary:

Designed a secure and scalable cloud infrastructure to support analytics workloads and backend services.

Highlights:

• Set up a production-ready infrastructure using EC2, with Application Load Balancers for routing and high availability.
• Configured multiple backend failover rules for resilient request processing.
• Used Azure DMS for near real-time data replication to downstream analytics tools like Power BI.
• Improved analytics data latency and enabled fail-safe access to production metrics.

4. Datalake Architecture for Automated Dashboarding

Role: Data Analytics Lead / FMCG SME

Industry: FMCG / Retail

Duration: October 2024 – Present

Tech Stack: AWS S3, AWS Glue, OData, Tableau,AWS Step Functions,AWS Lambda AWS Athena, Great Expectations

Summary:

Built a unified data platform using a multi-layered Datalake architecture to power automated business reporting.

Highlights:

• Ingested SAP data via OData APIs and processed external Excel inputs into the AWS S3 Raw layer.
• Built a structured Raw → Silver → Gold pipeline using AWS Glue to facilitate clean, optimized data transformations.
• Integrated Great Expectations (Euclidean framework) to ensure data integrity and validation at each stage.
• Delivered over 100 Tableau dashboards using more than 1000 curated tables, queried efficiently via AWS Athena.

5. AI-Powered Cardiac Emergency Triage System

Project Type: Personal Project

Industry: Healthcare

Duration: August 2025 – Present

Tech Stack: Python, BERT, LangChain, Whisper, OpenAI API, Twilio, FCM

Summary:

Designed and implemented a real-time AI-powered healthcare triage system to detect and respond to sudden cardiac emergencies using LLMs and NLP.

Highlights:

• Built a medical NER pipeline using BERT-based transformers for symptom extraction and emergency detection.
• Developed a hybrid rule + ML scoring engine to classify cardiac severity levels.
• Integrated real-time alerting through Twilio (SMS) and Firebase Cloud Messaging (push notifications).
• Enabled multilingual and voice input via Whisper and Google Speech-to-Text APIs.
• Reduced simulated emergency response time from 30+ minutes to under 5 minutes in pilot environments.