Ansh Kesarwani profile photo 1 on Tal

Ansh Kesarwani

SDE 1 @Meesho | Backend Developer Intern@WithWeb | CSE’25 at IIIT Bhopal | 4⭐at CodeChef

Bengaluru

3+ years of work experience

Meesho
Nov 202510 mosPresent

Bengaluru, Karnataka, India

Meesho
Nov 202510 mosPresent

Authored the root-cause analysis for a business-critical shipping-label and invoice service running at 99% of its 13 GB memory limit across all 40 servers, ranking three causes from Grafana metrics and proving 10 GB was reserved against only ~6 GB in use. Eliminated ~1.2–2 GB per server of kernel cache by migrating all temporary PDF writes to a memory-backed tmpfs mount (/tmp/myapp), coordinated with the DevOps team. Reduced memory footprint by ~3 GB per server (~120 GB fleet-wide) by right-sizing a thread pool from 1500 to 500 (~1,545 threads sat 98.5% idle), then cutting the JVM heap from 10 GB to 7 GB — re-measuring the peak before each cut. Engineered the data pipelines powering Valmo delivery-capacity planning, building Kafka consumers that ingest demand, hub capacity and route pricing into MongoDB, backed by 5 automated cron jobs. Restructured the Valmo route cost model to report last-mile hub cost separately from the rest of the route, giving the pricing team a clearer breakdown, and rolled it out hub by hub behind a Zookeeper-controlled list for instant rollback. Optimized Redis usage with batch prefetching of pincodes, reducing lookups from 50 to 1 per request and cutting datastore load by 50x, plus a cache-first pincode-to-Valmo-hub mapping with dynamic TTLs. Developed the end-to-end data flow for MSV (Minimum Serviceable Volume) — the minimum volume a Valmo route needs to be viable — automating its cron job, Kafka consumer and reporting event. Delivered 9 file upload flows spanning Valmo hubs, routes, pincode and supplier mappings, delivery-time buffers and slots, enabling the operations team to update this data themselves by uploading a file. Partnered with the Data Engineering team to publish 9 silver tables in Prism, enabling operations staff to monitor pipeline failures directly on Metabase dashboards.
Meesho
May 20257 mosNov 2025

Bengaluru, Karnataka, India

Withweb
Sep 20245 mosJan 2025

Remote

Created an API to handle 50,000+ product records with efficient pagination and search, reducing response time from 400 ms to 60 ms. Implemented OTP verification using the 2Factor API and integrated PhonePe payment processing for secure transactions. Deployed the Node.js application on AWS EC2, managed with PM2, ensuring continuous uptime and monitoring.
CODÂME, IIIT Bhopal
Aug 20221 yr 10 mosMay 2024

Bhopal, Madhya Pradesh, India

Google Cloud Community India
May 20224 mosAug 2022

Bhopal, Madhya Pradesh, India

Ansh’s AI Stack

claude
claude
cursor
cursor

Education

Indian Institute of Information Technology Bhopal

Bachelor of Technology, Computer Engineering

Indian Institute of Information Technology Bhopal

20214 yrs2025

Join Ansh and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Ansh Kesarwani on Tal

Headline

SDE 1 @Meesho | Backend Developer Intern@WithWeb | CSE’25 at IIIT Bhopal | 4⭐at CodeChef

About

As a passionate and driven Computer Science student at IIIT Bhopal, I have developed a solid foundation in backend development, data structures, and algorithms. I thrive on solving complex problems and delivering optimized, scalable solutions. Currently, I am working as a Backend Developer Intern at WithWeb, where I’ve gained hands-on experience deploying Node.js applications on AWS EC2, building robust APIs, and integrating secure OTP verification systems. I’ve successfully optimized a product API to handle 50k+ records, improving search performance from 400ms to 60ms. My academic projects further fuel my enthusiasm for building innovative solutions. From creating an AI-powered document search engine using Google Generative AI and FAISS, to developing a full-stack developer networking platform, I enjoy applying my technical skills to impactful projects. I am also an active competitive programmer, consistently performing in the top 2% on CodeChef and holding a 1600+ rating on LeetCode. Key Skills: Backend Development (Node.js, Express, MongoDB) Cloud Deployment (AWS EC2, PM2) API Development and Optimization Competitive Programming (CodeChef 4-star, LeetCode 1600+) Leadership and Team Management I’m always open to opportunities where I can leverage my technical skills and learn new technologies, especially in backend development and cloud infrastructure. Let’s connect and explore how I can contribute to your team! My Resume-https://bit.ly/3zEDpnK

Location

Bengaluru

AI stack

claude

cursor

Work experience

SDE 1 at Meesho, Nov 2025 - Present · 10 mos, Bengaluru, Karnataka, India

Software Engineer at Meesho, Nov 2025 - Present · 10 mos, Authored the root-cause analysis for a business-critical shipping-label and invoice service running at 99% of its 13 GB memory limit across all 40 servers, ranking three causes from Grafana metrics and proving 10 GB was reserved against only ~6 GB in use. Eliminated ~1.2–2 GB per server of kernel cache by migrating all temporary PDF writes to a memory-backed tmpfs mount (/tmp/myapp), coordinated with the DevOps team. Reduced memory footprint by ~3 GB per server (~120 GB fleet-wide) by right-sizing a thread pool from 1500 to 500 (~1,545 threads sat 98.5% idle), then cutting the JVM heap from 10 GB to 7 GB — re-measuring the peak before each cut. Engineered the data pipelines powering Valmo delivery-capacity planning, building Kafka consumers that ingest demand, hub capacity and route pricing into MongoDB, backed by 5 automated cron jobs. Restructured the Valmo route cost model to report last-mile hub cost separately from the rest of the route, giving the pricing team a clearer breakdown, and rolled it out hub by hub behind a Zookeeper-controlled list for instant rollback. Optimized Redis usage with batch prefetching of pincodes, reducing lookups from 50 to 1 per request and cutting datastore load by 50x, plus a cache-first pincode-to-Valmo-hub mapping with dynamic TTLs. Developed the end-to-end data flow for MSV (Minimum Serviceable Volume) — the minimum volume a Valmo route needs to be viable — automating its cron job, Kafka consumer and reporting event. Delivered 9 file upload flows spanning Valmo hubs, routes, pincode and supplier mappings, delivery-time buffers and slots, enabling the operations team to update this data themselves by uploading a file. Partnered with the Data Engineering team to publish 9 silver tables in Prism, enabling operations staff to monitor pipeline failures directly on Metabase dashboards.

SDE Trainee at Meesho, May 2025 - Nov 2025 · 7 mos, Bengaluru, Karnataka, India

Back End Developer at Withweb, Sep 2024 - Jan 2025 · 5 mos, Remote, Created an API to handle 50,000+ product records with efficient pagination and search, reducing response time from 400 ms to 60 ms. Implemented OTP verification using the 2Factor API and integrated PhonePe payment processing for secure transactions. Deployed the Node.js application on AWS EC2, managed with PM2, ensuring continuous uptime and monitoring.

Executive at CODÂME, IIIT Bhopal, Aug 2022 - May 2024 · 1 yr 10 mos, Bhopal, Madhya Pradesh, India

Google Cloud Facilitator Program at Google Cloud Community India, May 2022 - Aug 2022 · 4 mos, Bhopal, Madhya Pradesh, India

Education

Bachelor of Technology, Computer Engineering, Indian Institute of Information Technology Bhopal