Skip to content
View nkaushal99's full-sized avatar

Block or report nkaushal99

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
nkaushal99/README.md

Hey, I'm Nikhil Kaushal 👋

Software Engineer · Backend & AI Engineering
Building reliable, production-grade systems at the intersection of distributed backend engineering and applied AI.

LinkedIn · Email


About me

I'm a software engineer focused on turning complex, high-throughput workflows into dependable products. My work spans Java/Spring backend systems, event-driven architecture, and LLM-powered applications—from multi-agent orchestration to retrieval, evaluation, and operational reliability.

  • 🧠 Building agentic AI platforms, RAG pipelines, and LLM evaluation systems
  • ⚙️ Designing asynchronous, resilient workflows with Kafka, SQS, and AWS
  • 🗄️ Modelling scalable data systems with DynamoDB, SQL/NoSQL, and vector databases
  • 🎓 B.E. Computer Engineering, Thapar Institute of Engineering & Technology — CGPA 9.3

Impact

  • Built a production multi-agent orchestration platform handling 10K+ weekly requests, with routing, validation, guardrails, and multi-step workflow control.
  • Engineered an event-driven RAG ingestion platform for document parsing, chunking, vector indexing, and stateful failure recovery.
  • Developed an asynchronous LLM-as-Judge evaluation pipeline with versioned rubrics; cut evaluation cost by 50% with batch inference.
  • Scaled sentence-level TTS through batched parallel synthesis—3× throughput and 70% lower inference cost.
  • Designed typed SQS FIFO workflows and DynamoDB concurrency controls to preserve tenant ordering and prevent lost updates.

Tech I work with

Java Python Spring Boot AWS Kafka DynamoDB Docker

AI Engineering — AI Agents · LangChain · LangGraph · RAG · LLM-as-Judge · Vector Databases
Backend & Distributed Systems — Spring Boot · REST APIs · Microservices · Event-Driven Architecture · Kafka · Amazon SQS · State Machines
Cloud & Data — AWS · Lambda · S3 · DynamoDB · SQL · NoSQL · Docker

Selected projects

Project Focus
One RAG Retrieval-augmented generation and document intelligence
VibeSense Agent AI agent that grounds music recommendations in live heart-rate signals and preferences
Event-Driven Microservices Asynchronous backend patterns and distributed workflows
Realtime Chat Real-time messaging system design and implementation

Currently exploring

Multi-agent architectures · AI evaluation and observability · Context engineering · MCP · scalable AI infrastructure

Credentials

  • AWS Certified Developer – Associate (2026–2029)
  • AWS Certified Solutions Architect – Associate

Open to conversations about backend systems, applied AI, and building reliable developer platforms.

Pinned Loading

  1. sonic-muse sonic-muse Public

    Java 4

  2. docs-diff docs-diff Public

    Java 1