Shirsh Bansal

Machine Learning Engineer

Bengaluru, Karnataka, India13 yrs 8 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Expert in building scalable distributed systems.
Proficient in training large computer vision models.
Strong leadership in engineering team development.

Stackforce AI infers this person is a Machine Learning and Distributed Systems Engineer with a focus on AI solutions in E-commerce.

Contact

Skills

Core Skills

Computer VisionMachine LearningImage TaggingAlgorithm DevelopmentDistributed SystemsBackend DevelopmentSystem Development

Other Skills

image and video intelligencedeployment pipelinesenterprise image taggingdocument classificationCLIP-based vision transformershyperparameter tuningmachine learning algorithmsreal-time prediction servicereal-time predictioncontextual advertisingdata processingdistributed architectureproduct listingsfeed processingcatalog management

About

Machine Learning & Distributed Systems Engineer | Computer Vision | E-commerce I specialize in building large-scale distributed systems and machine learning solutions that power real-world applications. With experience in e-commerce and ML domains, I’ve worked extensively with high-throughput systems using HBase, Elasticsearch, MySQL, Kafka, PostgreSQL, and Redis—ensuring scalable and efficient data infrastructure. My expertise spans training and finetuning large computer vision models, from unimodal image classification to multimodal LVLMs for image captioning, and deploying ML models at scale, serving thousands of QPS with submillisecond latencies. I have hands-on experience in ML pipeline automation, distributed system architecture, and leading engineering teams in computer vision and deep learning. Passionate about image and video generation, I explore advancements in generative AI to push the boundaries of what’s possible in media and content creation. Always eager to discuss new challenges and innovations in AI, vision, and scalable systems—let’s connect!

Experience

13 yrs 8 mos

Total Experience

4 yrs 6 mos

Average Tenure

8 yrs

Current Experience

Adobe

3 roles

Machine Learning Engineer 5

Promoted

Feb 2024 – Present · 2 yrs 4 mos

In this role, I focused on enhancing image and video intelligence services through the development of advanced computer vision models. My role involved leading a team of engineers to create innovative solutions for asset performance prediction, training and optimizing multimodal models, and ensuring seamless deployment pipelines following the principles of responsible AI. I also mentored team members, fostering collaboration and technical growth within the group.

computer visionmachine learningimage and video intelligencedeployment pipelines

Machine Learning Engineer 4

Promoted

Jan 2021 – Jan 2024 · 3 yrs

I focused on enterprise image tagging and document classification, leveraging advanced machine learning techniques. I created a supervised model using CLIP-based vision transformers, optimizing it for performance and efficiency. Additionally, I implemented a scalable tagging service, achieving impressive latency and throughput. My work involved requirement analysis, dataset curation, hyperparameter tuning, and automation of custom model creation for enterprise documents following the principles of responsible AI.

enterprise image taggingdocument classificationCLIP-based vision transformershyperparameter tuningmachine learningimage tagging

Machine Learning Engineer 3

May 2018 – Jan 2021 · 2 yrs 8 mos

I contributed to the Ad services team by building a sophisticated pipeline for training machine learning algorithms. My focus was on utilizing user signals to predict business outcomes effectively. I successfully created a prediction service that operates in real-time, ensuring high performance and low latency. This role allowed me to deepen my expertise in machine learning and cloud technologies.

machine learning algorithmsreal-time prediction servicemachine learning

Media.net

Senior Platform Engineer

Jul 2017 – May 2018 · 10 mos · Mumbai Metropolitan Region · On-site

At Media.net, I played a pivotal role in enhancing contextual advertising through the development of an innovative algorithm that detects revenue-generating keywords related to web page content. This initiative aimed to significantly increase click-through rates and overall revenue. My responsibilities included processing vast amounts of data daily, ensuring the algorithm's efficiency and effectiveness in a fast-paced environment.

contextual advertisingalgorithm developmentdata processing

Flipkart

4 roles

Software Development Engineer 3

Promoted

Jul 2016 – Jul 2017 · 1 yr

I played a key role in enhancing the Seller Platform by developing robust systems for product listings and feed processing. My focus was on designing a distributed architecture that could efficiently handle millions of requests daily while maintaining data integrity and governance. I utilized advanced distributed technologies to implement solutions that significantly improved processing efficiency and offered real-time updates making sure they work under high concurrency, throguhput and provide low latency.

distributed architectureproduct listingsfeed processingdistributed systems

Software Development Engineer 2

Promoted

Jul 2014 – Jun 2016 · 1 yr 11 mos

I focussed on the Seller Platform by enhancing catalog management and optimizing pricing workflows. My role involved cross-team collaboration to ensure project success and foster a collaborative environment. I developed backend APIs and implemented monitoring and auditing processes to improve overall efficiency in various workflows like catalog, listing, pricing and order management.

catalog managementpricing workflowsbackend APIsbackend development

Software Development Engineer 1

Jul 2012 – Jun 2014 · 1 yr 11 mos

In this role at Flipkart.com, I contributed to the Catalog Management System Team by developing an audit framework and building a seller platform from the ground up. This platform successfully scaled to manage millions of listings, enhancing the efficiency of product creation and updates through tailored ingestion pipelines. My work focused on ensuring system reliability and improving seller experience.

audit frameworkseller platform developmentsystem development

Technology Intern

Jan 2012 – Jun 2012 · 5 mos

During my internship at Flipkart, I contributed to the Catalog Management System team by focusing on database migration to a new schema. This project involved migrating millions of book products while ensuring seamless functionality without downtime. Additionally, I played a key role in automating the catalog quality control workflow, which improved overall efficiency and audit capabilities.

database migrationcatalog quality control