Looking for a job

Kshitisha User

@kshitisha

Message

AI Engineer building production LLM systems with RAG and multi-agent orchestration.

India

Message

What I'm looking for

I’m looking for a role where I can build and deploy production-ready LLM/RAG systems end-to-end—designed for monitoring, evaluation, and scale—working with teams that care about quality gates, reliability, and real user impact.

I’m an AI engineer who ships production systems—not prototypes. I built and deployed RAG pipelines, multi-agent orchestrators, and LLM-powered services that handle real user traffic, and I’m comfortable taking an idea from script to a containerised, monitored production endpoint.

Experience

Work history, roles, and key accomplishments

AI Intern

AgilityAI

Aug 2025 - Nov 2025 (3 months)

Built and deployed a production RAG pipeline for automated test generation (CLAT, IPMAT, CUET) using LangChain and Gemini 2.5 Flash, reducing paper creation time from 40 hours to 30 minutes. Implemented an eval/diversity layer and automated quality gates with 60%+ coverage on critical generation paths.

Automated Testing LLM Evaluation Keyword Extraction Forbidden List Propagation Quality Gates Answer Distribution Validation Test Coverage

Data Science Intern

Celebal Technologies

Jun 2025 - Aug 2025 (2 months)

Benchmarked Random Forest, SVM, and XGBoost for a 4-class plant disease classifier and performed per-class failure analysis using confusion matrices. Automated the end-to-end ML pipeline (preprocessing to persistence), cutting turnaround time by 60% with zero manual handoffs.

Random Forest SVM XGBoost Plant Disease Classification Benchmarking Training Evaluation Reporting Model Persistence

Education

Degrees, certifications, and relevant coursework

University of Petroleum and Energy Studies (UPES)

Bachelor of Technology, Computer Science (AI/ML Specialisation)

2022 - 2026

Grade: CGPA: 7.7 / 10

Activities and societies: Advisor UPES CSA, Student Chapter

AI engineer who ships production systems — not prototypes. Built and deployed RAG pipelines, multi-agent orchestrators, and
LLM-powered services handling real user traffic. Comfortable taking an idea from script to containerised, monitored, production
endpoint.