Abhishek Kumar

Director of Engineering

Menlo Park, California, United States10 yrs 6 mos experience
Highly Stable

Key Highlights

  • Led critical product initiatives at Meta.
  • Grew teams significantly across multiple projects.
  • Delivered state-of-the-art ML initiatives.
Stackforce AI infers this person is a B2C Social Media expert with strong capabilities in Machine Learning and Product Development.

Contact

Skills

Core Skills

Machine LearningRecommendationsProduct DevelopmentFeed RankingSearch Ranking

Other Skills

AlgorithmsCC++Data AnalysisData StructuresDistributed SystemsEclipseGitJavaLinuxModelingMySQLPHPServletsTeam Leadership

About

Led multiple of Meta’s critical 0-1 products/initiatives to success and sustainability - Instagram Reels, Instagram Feed Recommendations, Instagram Labs, Facebook Stories, Facebook Marketplace Search, Facebook Search Whole Page Ranking and Facebook Trending. ~12 years of experience in Machine Learning, Search and Recommendation, delivering multiple SoTA ML initiatives, particularly focused on foundational models (LLMs) in Recommendations. Grew/hired 15+ Senior Managers (M2s), Senior Staff Engineers (IC7) and Distinguished Engineers (Engineering Director IC or IC8).

Experience

Instagram

Director of Engineering, Instagram Relevance

Jun 2021Present · 4 yrs 9 mos · Menlo Park, California, United States · On-site

  • 𝐋𝐞𝐝 𝐈𝐆 𝐑𝐞𝐞𝐥𝐬 𝐑𝐞𝐥𝐞𝐯𝐚𝐧𝐜𝐞(𝐢.𝐞. 𝐌𝐋+𝐈𝐧𝐟𝐫𝐚) 𝐬𝐢𝐧𝐜𝐞 𝐢𝐭𝐬 𝐧𝐞𝐚𝐫 𝐢𝐧𝐜𝐞𝐩𝐭𝐢𝐨𝐧
  • Brought Reels from being negative to IG’s engagement and retention to being the majority(50%+) contributor to IG’s year-over-year engagement and retention growth. Grew Reels from accounting for less than 5% of time-spent on IG, to 50%+. Achieved these during a span of 2 years.
  • Grew the team from 25 to 150+, including 4 Senior Managers, 7 Senior Staff Engineers and 1 Distinguished Engineer(Engineering Director)
  • 𝐋𝐞𝐝 𝐈𝐆 𝐅𝐞𝐞𝐝 𝐑𝐞𝐥𝐞𝐯𝐚𝐧𝐜𝐞 𝐚𝐧𝐝 𝐛𝐫𝐨𝐮𝐠𝐡𝐭 𝐭𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐨𝐧𝐚𝐥 𝐜𝐡𝐚𝐧𝐠𝐞𝐬 𝐢𝐧 𝐭𝐡𝐞 𝐩𝐫𝐨𝐝𝐮𝐜𝐭 𝐚𝐧𝐝 𝐬𝐲𝐬𝐭𝐞𝐦𝐬
  • Built IG Feed Recommendations from scratch and over 1.5 years establishing it as the majority
  • contributor to IG Feed’s year-over-year engagement and retention growth. Grew its footprint(impressions) on Feed from 0 to 30%.
  • Grew the team from 0 to 30. Hired/grew 2 Senior Managers and 2 Senior Staff Engineers.
  • Successfully delivered the largest unification of 2 relevance systems at Meta: IG Feed’s Recommendations and Connected Ranking - beating the deadline by 30%.
  • Drove merging of Feed Recommendations and Connected Feed teams into IG Feed Relevance, a team of 80+ ICs+EMs.
  • 𝐅𝐨𝐮𝐧𝐝𝐞𝐝 𝐈𝐆 𝐋𝐚𝐛𝐬 𝐑𝐞𝐥𝐞𝐯𝐚𝐧𝐜𝐞 𝐭𝐞𝐚𝐦 𝐚𝐧𝐝 𝐬𝐮𝐜𝐜𝐞𝐬𝐬𝐟𝐮𝐥𝐥𝐲 𝐢𝐧𝐜𝐮𝐛𝐚𝐭𝐞𝐝 𝐦𝐮𝐥𝐭𝐢𝐩𝐥𝐞 0-1 𝐏𝐫𝐨𝐝𝐮𝐜𝐭<>𝐌𝐋 𝐈𝐆 𝐛𝐞𝐭𝐬
  • Successfully launched multiple 0-1 IG big-bets with strong impact - Reposts, AI translation/dubbing, Friendly Feed, Photos-Carousels in Reels Tab.
  • Grew the team from 0 to 40 including 1 Senior Manager and 2 Senior Staff Engineers.
  • 𝐅𝐨𝐮𝐧𝐝𝐞𝐝 𝐭𝐡𝐞 𝐑𝐞𝐥𝐞𝐯𝐚𝐧𝐜𝐞 𝐂𝐨𝐫𝐞 𝐌𝐋 𝐌𝐨𝐝𝐞𝐥𝐢𝐧𝐠 𝐚𝐧𝐝 𝐃𝐞𝐥𝐢𝐯𝐞𝐫𝐲 𝐭𝐞𝐚𝐦𝐬 (𝐜𝐮𝐫𝐫𝐞𝐧𝐭𝐥𝐲 60+ 𝐜𝐨𝐦𝐛𝐢𝐧𝐞𝐝)
  • Delivered several SoTA ML initiatives successfully: Foundational models(LLMs), Longterm user interaction(sequence) modeling through Transformers, Content Exploration.
Machine LearningRecommendationsTeam Leadership

Facebook

2 roles

Senior Engineering Manager, Feed and Stories

Promoted

Jun 2018Jun 2021 · 3 yrs · Menlo Park, California, United States

  • 𝐈𝐧𝐢𝐭𝐢𝐚𝐭𝐞𝐝 𝐚𝐧𝐝 𝐥𝐞𝐝 𝐅𝐁’𝐬 𝐒𝐭𝐨𝐫𝐢𝐞𝐬 𝐑𝐚𝐧𝐤𝐢𝐧𝐠, 𝐞𝐬𝐭𝐚𝐛𝐥𝐢𝐬𝐡𝐢𝐧𝐠 𝐢𝐭 𝐚𝐬 𝐭𝐡𝐞 𝐥𝐚𝐫𝐠𝐞𝐬𝐭 𝐜𝐨𝐧𝐭𝐫𝐢𝐛𝐮𝐭𝐨𝐫 𝐭𝐨𝐰𝐚𝐫𝐝 𝐅𝐁 𝐒𝐭𝐨𝐫𝐢𝐞𝐬 𝐲𝐞𝐚𝐫-𝐨𝐯𝐞𝐫-𝐲𝐞𝐚𝐫 𝐠𝐫𝐨𝐰𝐭𝐡, 𝐚𝐧𝐝 𝐚 𝐤𝐞𝐲 𝐝𝐢𝐟𝐟𝐞𝐫𝐞𝐧𝐭𝐢𝐚𝐭𝐨𝐫 𝐜𝐨𝐦𝐩𝐚𝐫𝐞𝐝 𝐭𝐨 𝐒𝐧𝐚𝐩𝐜𝐡𝐚𝐭.
  • Grew the team from 0 to 20 over a period of 1.5 years.
  • Led FB’s Feed Ranking, delivering sustained impact - largest contributor to engagement on FB.
  • Founded FB’s Music Recommendation initiative and established it as an impactful lever for Creation - 15% of FB’s Creation growth for 3 years following its inception.
Feed RankingTeam LeadershipProduct Development

Staff Engineer / Engineering Manager, Facebook Search

Oct 2013Jun 2016 · 2 yrs 8 mos · Menlo Park

  • 𝐈𝐧𝐢𝐭𝐢𝐚𝐭𝐞𝐝 𝐚𝐧𝐝 𝐬𝐮𝐜𝐜𝐞𝐬𝐬𝐟𝐮𝐥𝐥𝐲 𝐝𝐞𝐥𝐢𝐯𝐞𝐫𝐞𝐝 𝐬𝐞𝐯𝐞𝐫𝐚𝐥 𝐒𝐞𝐚𝐫𝐜𝐡 𝐚𝐧𝐝 𝐫𝐞𝐥𝐚𝐭𝐞𝐝 𝐯𝐞𝐫𝐭𝐢𝐜𝐚𝐥𝐬
  • Built a quality Marketplace Search ranking singlehandedly in less than a week. This unblocked the launch of Marketplace and was a sustained foundation for ranking impact over the next 2 years.
  • Took on as the TL on FB’s Trending product. Grew its engagement by 40%+ over a period of 5 months, through a team of 4. Shipped Search’s first realtime and sparse heavy large ML model.
  • In addition, delivered impactful launches on ranking and infra for almost all Search verticals - entities, videos, photos, places, news.
  • 𝐓𝐋 𝐨𝐧 𝐒𝐞𝐚𝐫𝐜𝐡’𝐬 𝐖𝐡𝐨𝐥𝐞 𝐏𝐚𝐠𝐞 𝐑𝐚𝐧𝐤𝐢𝐧𝐠 (8 𝐈𝐂𝐬)
  • blending and ranking Search results from 10+ content types.
  • 𝐁𝐮𝐢𝐥𝐭 𝐅𝐁 𝐒𝐞𝐚𝐫𝐜𝐡’𝐬 𝐞𝐧𝐠𝐚𝐠𝐞𝐦𝐞𝐧𝐭 𝐜𝐨𝐮𝐧𝐭𝐞𝐫 𝐩𝐥𝐚𝐭𝐟𝐨𝐫𝐦.
Search RankingTeam LeadershipMachine Learning

Education

Indian Institute of Technology (Banaras Hindu University), Varanasi

Masters of Technology — Computer Science and Engineering

Jan 2012Jan 2013

Indian Institute of Technology (Banaras Hindu University), Varanasi

Bachelor of Technology — Computer Science and Engineering

Jan 2008Jan 2012

Stackforce found 100+ more professionals with Machine Learning & Recommendations

Explore similar profiles based on matching skills and experience