Abhijith CLead AI Engineer
Work Experience

Lead AI Engineer, Data Sciences (Staff MLE Scope)

Target
  • •Led migration of Offer Personalization Engine to GCP, defining system architecture and scaling strategy
    • 05/2024 – presentBangalore
  • •Designed end-to-end architecture for the Next Best Action (NBA) system, including pipelines, contracts, and cross-team integrations
  • •Built state metadata system as source of truth for NBA pipelines, enabling reliable decisioning and state transitions at scale
  • •Implemented multi-armed bandit (MAB) models for category-level optimization, improving category-level decisioning for personalized offers
  • •Defined and drove alignment on a converged MLOps “Golden Path” (GCP stack), reducing tooling fragmentation and standardizing ML workflows across teams
  • •Influencing architecture and design of Target’s ML Platform (feature store, training infra, serving, monitoring)
  • •Mentored engineers on ML platform development and system design (e.g., pipelines framework, campaign systems)

    • Staff Machine Learning Engineer 1

    Glance, InMobi
  • •Architected real-time recommendation and content serving system using PySpark Structured Streaming + Kafka + gRPC [blog⁠]
    • 01/2024 – 05/2024Bangalore
  • •Led team delivering recommendations for 100M+ DAUs with <500ms latency
  • •Applied bandits (Thompson Sampling) improving engagement by ~20%
  • •Reduced cloud costs by ~$50K/month via infra and workload optimizations
  • •Led edge ML initiatives (TFLite) for on-device inference
  • •Established production-grade ML practices (monitoring, logging, performance testing) for large-scale serving systems on GCP [blog⁠]
  • •Designed ANN-based candidate generation systems for dynamic feeds

    • Machine Learning Engineer 1, 2, 3

    Glance, InMobi
  • •Built automated content creation platform generating 95%+ of published content
    • 08/2019 – 12/2023Bangalore
  • •Developed ranking systems for content selection and optimization
  • •Designed real-time streaming pipelines (Azure EventHub, GCP Pub/Sub) powering recommendation systems
  • •Built CV/NLP systems for deduplication, ranking, editing, and enrichment
  • •Developed video transformation system (landscape → portrait) preserving salient regions

    • Data Scientist

    InMobi
  • •Built CV models for ad quality optimization and super-resolution
    • 2018 – 2019Bangalore
  • •Developed "Creative Insights" system for ad performance evaluation
  • •Improved moderation efficiency via deduplication (SIFT, Siamese Nets) and classification systems
    • Education

    Bachelor of Engineering, CS

    Sir M. Visvesvaraya Institute of Technology
    2014 – 2018Bangalore

    Master of Science, CS (ML Specialisation)

    Georgia Institute of Technology
    2022 – 2024
    Achievements

    Mastermind in Data Science

    InMobi

    Multiple Quaterly Best Performer Awards

    InMobi + Glance

    Honorable mention, ACM ICPC 2016 & 2017

    ACM ICPC

    Google Summer of Code, with Python Software Foundation

    Google
    Publications

    Deduplication of Advertisement Assets Using Deep Learning Ensembles⁠

    19th IEEE International Conference On Machine Learning And Applications

    Provided is a system and method for digital content generation and recommendation.

    Other Projects

    GSoC 2017 project under Italian Mars Society (sub org. Under Python Software Foundation)

  • •This project mainly revolves around configuring biometric signal sensors, using Hexoskin smart shirt
  • •Integrating the the smart shirt (the sensors) with Tango-Controls Server, SQLAlchemy ORM for the database

  • •An approach to use CapsNet as a discriminator in Generative Adversarial Network and demonstrate its appli- cation using semantic inpainting on MNIST and face images.