Chirag Lalwani

Data Scientist

Bengaluru, Karnataka, India8 yrs 9 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Led end-to-end data tech for impactful products.
Built Learning to Rank product generating $1.5M annually.
Published research papers and provisional filing.

Stackforce AI infers this person is a Data Scientist with strong expertise in SaaS and AI-driven solutions.

Contact

Skills

Core Skills

Deep LearningSearch Engine RankingData MiningMedical ImagingMachine LearningPredictive AnalyticsData AnalyticsData Analysis

Other Skills

Natural Language Processing (NLP)Agile MethodologiesdatabricksApache SparkComputer VisionTensorFlowGitMLflowAmazon Web Services (AWS)Problem SolvingPython (Programming Language)Artificial Neural NetworksPySparkNLPAWS SageMaker

About

With significant experience, I manage to contributed in Data Science domain by being part of multiple industry verticals, bringing additional exposure from working with an early-stage startup. Academically, published couple of research papers and a provisional filing. I have exposure in leading end-to-end data tech for customer facing and internal products bringing notable impact to business. Some of recent hustling involves building & leading gallery ranking from scratch and improving former gallery search. Completed B.tech with merit, majored in computer science.

Experience

8 yrs 9 mos

Total Experience

1 yr 6 mos

Average Tenure

2 yrs 6 mos

Current Experience

Apple

Data Scientist

Dec 2023 – Present · 2 yrs 5 mos · Bengaluru, Karnataka, India · Hybrid

Vistaprint

Senior Data Scientist

Sep 2021 – Dec 2023 · 2 yrs 3 mos · Bengaluru, Karnataka, India

➢ Built a Learning to Rank (LTR) product through Deep Learning recommender algorithm from scratch for
all business card galleries across all regions end-to-end (from scalable data wrangling using PySpark &
Petastorm to moving in real-time production env using MLflow and Databricks jobs) bringing $1.5 M
annualized VGP to business.
➢ Improved gallery search product by 7% through multi-model encoder embeddings which intends to map
user’s search on website with designs images and meta-data in runtime through Solr knn vector search;
semantically. Fine-tuned and deployed ViT model by through custom flavored MLflow and AWS
SageMaker under automated pipelined platform with validations.
➢ Successfully Led Gallery Design Findability technical roadmap by composing infrastructure, priorities and
versioning; in close collaboration with Product Owner. Made mark in vista’s year deck in addition to vista
wide technical blog.
➢ Built multiple internal tooling with Analytics consultancy to stakeholders and other teams. An example – By
automatically detecting and separating inspirational design templates in gallery from the plain ones by
OpenCV, PySpark and supervised ensemble classification algorithm across all locales processing more than
1M entities.
➢ Helped in growth of multiple Data Scientists and Data Engineers by providing technical mentorship, sharing
best practices, offering tailored guidance under center of excellence opportunity.
➢ Collaborated in keeping Business in-line with industry state-of-art by Researching and building successfully
proof-of-concept to automatically Animate the designs through U-Net and automatic Design generation from
user’s description through advance variational autoencoders.