×
0 followers0 following
Jakka Shashank
Telangana, India0 followers0 following
- About
- High School
- Bachelor's
- Master's
- Target
Male
Telangana, India
Work ExperienceLead, Machine learning Enginner
Sharechat - Full Time
Jan 4, 2024 - Present · 2 yrs 4 mos
Karnataka, India
Working to build large scale notification recommendation system for India's largest short video streaming app Moj.
Advanced Analytics Intern
Digital.ai - Internship
May 1, 2019 - Sep 1, 2019 · 4 mos
Karnataka, India
Developed an advanced analytics workflow utilizing a decision tree model for predicting risk of change failure. Implemented clustering algorithms and backtracking techniques to analyze root causes of failures, enabling the recommendation of optimized parameter ranges to significantly reduce failure rates.
Computer Vision Research Intern
Curl analytics - Internship
Jan 1, 2019 - Apr 1, 2019 · 3 mos
Karnataka, India
Explored several detection models for real time road inspection like detecting cracks, potholes for the government of Andhra Pradesh
Improve the state of the art models for semantic segmentation like FCN to support semi supervised learning and also with conditional random fields for enhanced segmentation
Data Science Intern
Yodlee - Internship
May 1, 2018 - Jul 1, 2018 · 2 mos
Karnataka, India
Determining the feasibility of predicting Income classes of US customers from the large transactional data.
Refining and laying down new improved rules for calculating the ground truth.
Increased the features from 36 to 107 and improved the current model accuracy by 16% (91% in production).
Also extended the project for predicting more granular income classes.
Computer Vision Intern
TRI3D - Internship
May 1, 2017 - Jul 1, 2017 · 2 mos
Telangana, India
Built an ML based add-on for Blender to learn to add springs after positioning cloth patterns on 3D human mesh to perform cloth simulation.
Extract the garment from the image by region growing algorithms.
Saved about 10-20 minutes of manual work to digitalise all the parts of the garment which were already digitalised by building a hashing based image matcher.
Machine learning Enginner
Mercari - Full Time
Oct 1, 2019 - Mar 31, 2022 · 2 yrs 6 mos
Tokyo, Japan
Enhanced Mercari's Price Suggestion Service, aiding 30M users daily with AI-driven pricing for over 3M listings. Leveraged DeepFM and Multi-Armed Bandits for scalable, real-time price optimization. Additionally, refined A/B testing methodologies and spearheaded initiatives for substantial improvements in key business metrics.
Machine learning Enginner
SmartNews - Full Time
Mar 1, 2022 - Dec 31, 2023 · 1 yr 10 mos
Tokyo, Japan
Advanced the candidate generation capabilities at SmartNews by developing and training Two-Tower and various deep neural network (DNN) based models. These models were designed to capture complex user-article interactions, enhancing the recommendation system's ability to accurately predict and personalize news article rankings, leading to a significant uptick in user engagement metrics.
ProjectsML driven Smart Pricing Optimization System
Role - Machine learning Enginner
for Industrial
Apr 1, 2020 - May 1, 2020
I led the development of the Smart Pricing Optimization System aimed at transforming the online marketplace experience. It is a deep neural network (DNN)-based model designed to dynamically adjust listing prices on the fly. Addressing a prevalent user challenge, the model intelligently prompts sellers to reconsider their pricing strategy when attempting to cancel a listing. By analyzing listings with high potential yet hindered by suboptimal pricing or premature cancellation, the system effectively reduces cancellation rates and enhances listing longevity. Within a month of implementation, Smart price boosted selling efficiency, generating over 150,000 additional sales and $8 million in revenue. This innovation not only improved the user experience by simplifying the pricing adjustment process but also underscored the power of machine learning in creating a more fluid and responsive online selling environment
Dynamic Pricing Optimisation Engine with Personalised Recommendations
Role - Machine learning Enginner
for Industrial
Mar 1, 2021 - Aug 31, 2021
In response to the challenges posed by the dynamic post-COVID market environment, which rendered the existing DeepFM-based price recommendation system at Mercari less effective due to its inability to adjust quickly to rapid changes in pricing behaviors, I spearheaded the development of a novel solution. Recognizing the limitations of frequent retraining of the DeepFM model and the extensive overhaul required for real-time training data pipelines, I proposed and created a proof of concept for a more agile and adaptive approach. By integrating a Multi-Armed Bandits (MAB) based reinforcement learning model atop the existing DeepFM framework, this solution was designed to capture shifts in pricing behavior in real-time. This hybrid model significantly enhanced the platform's ability to offer better, more personalized price suggestions, leading to a +6% increase in average Gross Merchandise Value (GMV) and a +1.2% improvement in Sell Through Rate (STR), alongside faster selling speeds for listings. This initiative not only demonstrated the feasibility of augmenting traditional models with reinforcement learning for improved responsiveness but also established a scalable, efficient pathway to adapt to rapidly changing market conditions without necessitating major infrastructural changes.
Optimized Supply Chain Forecasting System
Role - Machine learning Enginner
for Industrial
Feb 10, 2018 - Feb 21, 2018
In a high-stakes international hackathon, I developed a hierarchical time series forecasting model to predict demand for 81 products across 16 Indian regions, utilizing data from 3 McKinsey manufacturing plants. I innovated the supply chain management process by formulating the sequencing problem with Hidden Markov Chains and the scheduling issue as an Integer Constraint Programming problem, moving beyond traditional greedy approaches. This methodology not only optimized the schedule, sequence, and distribution of products efficiently but also secured me the first prize, outperforming numerous Kaggle grandmasters. My work was recognized at the IEOR Day 2018 at IIT Bombay, where I was invited as a keynote speaker to present my solution, highlighting its impact on supply chain optimization and showcasing my analytical prowess and problem-solving capabilities.
Research PapersJakka Shashank has not added any research paper yet.
VolunteeringJakka Shashank has not added any volunteering detail yet.