Leading Apollo's Machine Learning, Data, and several vertical products (Prospecting, Sales Engagement, Plays, Actions, Meetings, and Conversations Intelligence). Currently leading an org of 100+ people with ~12 teams spanning across APAC (India, Bangladesh, Philippines, Vietnam), USA, Canada, and Poland.

Apollo.io is a leading provider of sales intelligence products. Backed by Sequoia Capital and Bain Capital Ventures, it became the first unicorn in the sales domain in 2023 with a valuation of $1.6B. I lead all things data and machine learning at Apollo, including all full-stack verticals that utilize data and ML. My org has 10+ teams led by managers, senior managers, staff engineers, and principal engineers, with a total size of 75+ people. These teams are spread across 5+ countries and operate in a fully remote setting.My org’s horizontal charter includes Data Platform, Data Engineering, ML Platform (including LLM Ops), ML Engineering, and Search Relevance. My vertical charter includes Apollo’s foundational products like Data offerings, Prospecting (search engine), Sales Engagement, and Plays.I am also leading Apollo’s Engineering Hiring program and have successfully hired more than 150 engineers on the Engineering team, a growth of ~700% since Jan 2022.

Fika Ventures is an early-stage venture capital firm focused on B2B software startups, with an emphasis on pre-Series A companies. Our expertise spans vertical SaaS, fintech, commerce enablement, healthcare, and marketplaces, particularly those harnessing AI to drive innovation. We pride ourselves on a hands-on, collaborative approach, viewing ourselves as true partners in building your business—not the other way around.

Storepath helped eCommerce sellers in expanding their business to multiple channels and countries. Our AI and data-driven approach allowed us to automate most aspects of the eCommerce business so sellers, brands, and merchants can focus on building and improving their products.Storepath raised $4M in funding from Fika Ventures, Neo (Ali Partovi's VC firm), Act One Ventures, and other angels.I led the technology (and whatever else that came my way) at Storepath. We launched approximately one big product per month and experimented rapidly to gather quick feedback from customers. Some of our notable products were: AI-powered seller tools (Copy Creator, Keyword Researcher, Listing Optimizer, Merchant Portal), an on-demand marketplace (webapp and mobile app) to connect eCommerce brands with freelancers, a Shopify app for automated Copy Creation using OpenAI's GPT-3 models, and an ML model to predict the valuation of eCommerce businesses.

Led multiple full-stack teams and initiatives, launching multiple highly impactful projects for student Admissions and Learning. -- Introduced and set up various engineering processes like Scrum, Oncall, and Engineering Onboarding while improved many others like Hiring, Roadmap, Tech Planning, and Engineering Documentation.-- Scaled up engineering teams and established good cross-functional working relationships that increased the overall productivity of the company. Handled build vs buy decisions for various products.-- Led initiatives responsible for 200% growth in student enrollments and a 50% drop in student withdrawals.-- Launched the Growth Experimentation program responsible for 15% increase in enrollment conversions.

I led the Data & Machine Learning Platform team at Amazon Music. My team built tools and platforms to help Scientists and Engineers access, process, compute, experiment, store & share data and build scheduled machine learning and other complex data workflows in production environments. Our platforms ran at Amazon scale processing billions of events per day.My team was closely aligned with Machine Learning, BI, and Product teams and we contributed to multiple cross-team programs. We owned the end-to-end batch data infrastructure at Amazon Music which includes the behavioral data ingestion ETLs, Redshift based data warehouse, Tableau based reporting, and Spark based data processing workflows. We also built platforms for large scale machine learning experimentation, workflow orchestration, and data accessibility. Being the custodians of Amazon Music customers' data, we were responsible for building security and privacy features in our platforms. Our products had high adoption rate and powered crucial features on the Amazon Music app, including personalization and music discovery.

I started the Data Science Platform at Amazon Music. Strategized the ML Platforms at Amazon Music, aligning closely with the central Amazon and AWS ML Platform teams. Worked with AWS Glue and AWS SageMaker teams during the early days of these products and built similar platforms at Amazon Music.

Led the development of various search relevance and query understanding initiatives, including, category correction, product type identification in search queries, and building out the backend data infrastructure. These products were launched to millions of users and handled millions of search queries per day.

Built features and integrations in Amazon's Ad Exchange, primarily serving contextual ads. The service processed billions of requests per day. Worked under tight SLA, throughput, and latency requirements.

Work included Named Entity Recognition, crowd-sourced data labeling using Amazon Mechanical Turk, Sentiment Analysis, Ad-targeting, classification, topic extraction, tagging, ETL processing, unstructured text annotator pipelines, and hadoop based systems. Contributed to open source (UIMA, ClearTK, Chukwa), including up-gradation of Concept Mapper UIMA annotator to version 2.4.

Teaching assistant for "Java and J2EE Programming".

Pursued Masters in Computational Data Science (formerly, VLIS).

Implemented the boost-Clustering algorithm using K-means as the base clustering algorithm, on the GameSpot (www.gamespot.com) data of CBS. Applied topic modeling using LDA (from Mahout library) to reduce the dimensionality of the input documents and gave this as input to the Boost-cluster algorithm.

Built an Open Source generic Recommendation System in Ruby called Prastava based on a hybrid algorithm of content based filtering and collaborative filtering. Project Guide - Dr. Min-Yen Kan, Assistant Professor, NUSTeam Size - 2 (Himanshu Gahlot and Tarun Kumar)Web Service Link - www.aye.comp.nus.edu.sg/~himanshu/pr/Project hosted at Github - https://github.com/hgahlot/Prastava

Worked as an intern on two projects : "Named Entity Recognition for Hindi" and "Shallow Parsing for Hindi". These projects were intended to be part of the automatic annotation system of the Hindi language processing tool "Sanchay". Obtained state of the art results in both tasks.Team Size - 2 (Himanshu Gahlot and Awaghad Ashish Krishnarao)Project Guide - Anil Kumar Singh, Research Scholar, IIIT Hyderabad