Looking for a job

Anushka04 User

@anushka04user

Message

I build production voice agents, RAG systems, and multi-agent AI for healthcare.

India

Message

What I'm looking for

I'm looking to build reliable production AI systems, especially conversational agents, voice experiences, RAG, and multi-agent workflows where grounded outputs, evaluation, and operational impact matter.

I've built patient-facing conversational AI at Storyline Health, owning RAG retrieval, LangGraph agent orchestration, MCP tool integration, and real-time voice experiences for a behavioral telehealth platform.

At Thinking Stack, I deployed local reranker inference, built healthcare and insurance intelligence systems that reduced manual review by about 73%, and improved RAG answer accuracy by 96% through better document structuring, metadata, and retrieval guardrails.

I also build low-latency voice agents and evaluated multi-agent RAG systems, using Python, FastAPI, LiveKit, OpenAI, LangGraph, and production LLM evaluation tooling to make AI outputs reliable and grounded.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer

Current

Storyline Health AI

Mar 2026 - Present (6 months)

Built patient-facing conversational AI for a behavioral telehealth platform, owning RAG retrieval, agent orchestration, and tool integration over MCP. Worked on real-time voice interaction and benchmarked reranker models for retrieval quality.

RAG LangGraph MCP LiveKit VAPI Reranking Evaluation Python fastAPI

AI Engineer

Thinking Stack

Feb 2025 - Mar 2026 (1 year 1 month)

Set up vLLM-based serving to benchmark reranker candidates on CPT/ICD code prediction and deployed the selected model to ONNX Runtime for local inference. Owned system reliability, guardrails, and governance.

VLLM ONNX Runtime Reranking Benchmarking LLM Evaluation GuardRails Reliability Python

AI Engineer

AIQWIP

Sep 2024 - Dec 2024 (3 months)

Built a real-time voice agent on LiveKit, the OpenAI realtime API, and Django, with multi-accent speech recognition around 92%. Debugged the live audio streaming pipeline to bring latency under 500ms, cutting query resolution time by ~35%.

LiveKit OpenAI API Django WebRTC Speech Recognition Python