Senior Backend Engineer AI, Retrieval & Microservices

SHIKHAR TIWARI

New York, NY *******.**********@*****.*** 201-***-****

LinkedIn: linkedin.com/in/shikhar-tiwari-4a9ba7142 Github: github.com/shkrwnd Leetcode: leetcode.com/u/shkrwnd Senior Software Engineer with 7+ years of experience building distributed backend and full-stack systems, including AI-powered retrieval, personalization, and recommendation platforms.

Proven ownership of production systems across enterprise, e-commerce, and startup environments. EDUCATION

Rutgers University- Newark, NJ Dec 2025

Master of Science (MS) in Information Technology

Jaypee Institute Of Information Technology May 2018 Bachelor of Engineering in Computer Science & Technology SKILLS & CERTIFICATIONS

• Languages: Python, Java, TypeScript, JavaScript

• Frontend: React, Angular

• Backend: FastAPI, Spring Boot, Node.js, Ruby on Rails, Laravel

• Data & Messaging: PostgreSQL, MongoDB, Redis, Kafka, Elasticsearch, Neo4j, Milvus

• Cloud & AI: AWS, GCP, Azure, Docker, Kubernetes, PyTorch, TensorFlow, Retrieval Augmented Generation (RAG) PROFESSIONAL EXPERIENCE

Software Developer 3 Aug 2023 - Dec 2024

Walmart Global Tech

• Owned the design and development of backend microservices powering an internal enterprise AI assistant (“Cassidy”), supporting secure question answering over sensitive corporate data with fine-grained access controls and compliance requirements

• Built a Kafka-based ingestion and embedding pipeline to process data from multiple enterprise systems, generate vector embeddings, and persist them in a vector database to enable low-latency semantic retrieval.

• Designed service integrations and automated workflows using ServiceNow and Azure, enabling controlled onboarding of organizations, administrative units, and LLM access with auditable approvals.

• Ensured compliance-ready deployment of AI services by enforcing data isolation, authorization boundaries, and auditability across enterprise systems. Co Founder (Technical) Jan 2024 – Oct 2024

Stealth Startup (AI Recommendation System)

• Designed and implemented a production-grade recommendation platform combining two-tower retrieval models with LLM-based reasoning.

• Built a RAG-based personalization pipeline supporting multi-source ingestion, embedding generation, retrieval, and LLM-driven re-ranking, designed for plug-and-play integration into third-party applications.

• Developed a customer-facing control plane for data onboarding, pipeline configuration, A/B experimentation, and performance monitoring, enabling rapid iteration and evaluation of recommendation strategies.

• Implemented billing, plan management, and experiment tracking, supporting usage-based pricing and systematic comparison of model variants and retrieval strategies. Senior Software Developer Jan 2022 - Aug 2023

Pharmeasy

• Designed and built a dynamic page-layout rendering service enabling real-time personalization of products, offers, and UI widgets based on user context and behavior signals.

• Architected and scaled an ad-serving and recommendation pipeline for cross-sell and upsell use cases, contributing 1Cr+ in monthly incremental revenue.

• Implemented graph-based product discovery using Neo4j to model product hierarchies and relationships, enabling complex filtering and low-latency traversal queries at scale.

• Participated in architectural discussions and peer reviews to improve system scalability and maintainability. Software Developer Nov 2020 - Jan 2022

Shiprocket

• Led performance optimization of a high-volume communication platform delivering WhatsApp, email, and SMS notifications at scale, improving throughput and system stability.

• Drove API and database optimizations across a seller-facing monolith, reducing request latency and improving overall service responsiveness.

• Built an event-driven marketing data pipeline to capture and process user behaviour signals for targeted communication and analytics use cases.

• Hardened billing and payment execution paths to ensure transactional correctness under high concurrency.

• Developed scalable search infrastructure using Elasticsearch to enable low-latency access to orders and customer data. Software Engineer June 2018 - Aug 2020

Real Time Data Services

• Designed and implemented usage-based billing logic for VoIP calling services, supporting multiple payment plans and accurate metering of call activity.

• Built a web-based management dashboard for configuring and monitoring VoIP services including IVR flows, outbound call campaigns, and call routing rules.

• Integrated AWS S3 and Amazon Polly (Text-to-Speech) to generate, store, and manage audio recordings used dynamically within call flow pipelines.

• Contributed to performance optimization and reliability improvements across telephony pipelines by identifying bottlenecks and hardening critical execution paths. RELEVANT PROJECTS

Deep Reinforcement learning in Distributed dynamic spectrum access

• Tools: Python, TensorFlow

• Using Deep Recurrent Q learning for multi-agent co-operation in distributed dynamic spectrum access. (https://github.com/shkrwnd/Deep-Reinforcement-Learning-for-Dynamic- Spectrum-Access +96 forks and 244 stars)

Graph RAG to improve Information Retrieval in LLMs

● Tools: Langchain, Graph neural network, Milvus DB, Python, S3

● Built a graph-augmented RAG pipeline using SBERT, Graph Neural Network (GNN) based score propagation, and LLM answer generation (LangChain, Milvus).

● Improved retrieval quality over vanilla vector-based RAG on multi-hop queries. Improved retrieval quality over vanilla vector-based RAG on multi-hop queries. CERTIFICATIONS

● Deep Learning Nanodegree (Udacity)

● Android App Development

Contact this candidate