Aritra Biswas

Aritra Biswas

Senior Manager, AI Engineering

AB InBev

About me

I’m Aritra Biswas, a Senior Manager of Analytics at AB InBev and an independent researcher based in Bangalore. My work brings together statistics, machine learning, and software engineering, with a focus on turning sound methods into useful systems.

Outside work, I enjoy reading, cooking, and exploring new cuisines. Travel has taken me to 🇵🇱, 🇫🇷, 🇹🇭, 🇲🇾, 🇸🇬, 🇳🇱, 🇧🇪, 🇧🇷, 🇦🇪, and 🇻🇳, often with food as the highlight.

I write about practical lessons from data and technology, from statistical methods to the engineering choices that make systems dependable.

Download my resumé.

Interests
  • Machine Learning Engineering
  • Architecture
  • MLOps & Observability
  • Software Development
Education

  • M.Sc. in Statistics, 2017

    University of Delhi

  • B.Sc. in Statistics, 2015

    Presidency College

Experience

 
 
 
 
 
AB InBev
Senior Manager Analytics
AB InBev
Oct 2023 – Present Bangalore, Karnataka, India
  • Generative AI
  • Applied Machine Learning
  • Software Development
 
 
 
 
 
AB InBev
Senior Machine Learning Engineer
AB InBev
Sep 2022 – Sep 2023 Bangalore, Karnataka, India
  • Recommendation Systems
  • Distributed Computing
 
 
 
 
 
AB InBev
Senior Data Scientist
AB InBev
Sep 2021 – Aug 2022 Bangalore, Karnataka, India
  • Budget Optimization
  • Lift Measurement
 
 
 
 
 
AB InBev
Data Scientist
AB InBev
May 2020 – Aug 2021 Bangalore, Karnataka, India
  • Marketing Mix Modeling (MMM)
  • Software Development
 
 
 
 
 
The Nielsen Company
Data Scientist
The Nielsen Company
Jul 2018 – Apr 2020 Bangalore, Karnataka, India
  • Multi-touch Attribution (MTA)
  • Software Development
 
 
 
 
 
The Nielsen Company
Associate Data Scientist
The Nielsen Company
May 2017 – Jun 2018 Bangalore, Karnataka, India
  • Market Mix Modeling (MMM)
  • Scenario Planning
  • Budget Simulator

Recent Posts

FastAPI Resiliency: Circuit Breakers, Rate Limiting, and External API Management
Let’s assume that we are building a REST API service that depends on third-party APIs. To make the example more realistic, let’s say we are building a RAG system that uses Tavily (a web search API) and LLMs.
Mar 11, 2025 11 min read programming
FastAPI Resiliency: Circuit Breakers, Rate Limiting, and External API Management
On LLM reproducibility
One of the major issue with LLM output is its consistency and reproducibility. Here is kind of a hacky work around to make output from a LLM consistent (kind of). Also, a few observation regarding the same.
Mar 9, 2025 7 min read programming
On LLM reproducibility
Exponential Smoothing using Scikit-Learn wrapper & statsmodels
As we discussed in previous post, in this series we will be working on different type of regression problem and will try to parse them as sklearn model objects. Here are we will be working with ExponentialSmoothing from statsmodels library.
Dec 20, 2022 3 min read programming
Exponential Smoothing using Scikit-Learn wrapper & statsmodels
LASSO using Scikit-Learn wrapper & CVXPY
Those who are working in ML space of sometime must be aware that there are multiple python libraries out there which support different algorithms/models. When you are trying to implement best model it is not possible to use just a single library or even a single language and compute infra.
Dec 18, 2022 7 min read programming
LASSO using Scikit-Learn wrapper & CVXPY
Avoid relative path import hell in python
Notes on using Poetry to manage Python dependencies, project metadata, virtual environments, and package publishing.
Dec 18, 2021 2 min read programming
Avoid relative path import hell in python
See all posts

Contact