Headline

AI Engineer | Generative AI & LLM Systems | LangChain · LangGraph · RAG · Multi-Agent AI | Python · AWS · OpenAI | Production AI at Scale

About

I build AI systems that go from whiteboard to production — and I measure success in accuracy rates, latency cuts, and hours of human work eliminated. I’m an AI Engineer at NeuralWaves, where I design and ship production-grade AI systems — from LLM-powered agents to content moderation pipelines. I sit at the intersection of applied research and real-world engineering, turning prototypes into systems that handle real traffic, real users, and real SLAs. At NeuralWaves, I’ve shipped: • An AI Interview Bot (OpenAI + Gemini APIs) handling 500+ monthly sessions with real-time dynamic evaluation and a 40% boost in backend response efficiency • A content moderation system that hit 95% accuracy and cut manual review workload by 70% — still running in production • A multi-agent AI Career Copilot that routes queries across specialized sub-agents (Job Hunter, Interview Prep, Resume AI) using LangGraph + Claude’s tool-use API, with <3s average response time. My stack, end to end: LangChain · LangGraph · RAG · LLM Fine-Tuning (Llama2, Gemma) → Python · Java Spring Boot · React.js → AWS (EC2 · S3 · Lambda · Bedrock) · Docker What separates me: I don’t just prototype AI — I architect it for real traffic, debug it at 2 AM, and optimize it until the latency and cost numbers are defensible. What I’m looking for / building next M.Tech in Artificial Intelligence | GH Raisoni University 📌 Currently exploring: Applied AI Engineer, AI/ML Engineer, GenAI Engineer, Forward Deployed Engineer roles in Pune · Mumbai (on-site, remote, or hybrid) 🔗 GitHub: github.com/stuti-meshram | Open to connecting with builders and AI teams.

Location

Pune District

AI stack

antigravity

claude

Work experience

AI Engineer at Neuralwaves-systems, Dec 2023 - Present · 2 yrs 8 mos, Pune District, Maharashtra, India, AI Interview Bot — Core System Architected an LLM-powered interview simulation platform integrating OpenAI GPT-4 and Google Gemini APIs, supporting 500+ monthly evaluation sessions with dynamic question generation and real-time answer scoring Built the full backend in Python + Spring Boot; optimized API orchestration and async processing pipelines, resulting in a 40% improvement in response throughput Reduced system latency by 35% through query optimization, REST API refactoring, and database indexing across microservices architecture Designed a modular LLM integration layer enabling hot-swappable AI models (GPT-4, Gemini, future models) with zero code changes — reducing new model integration time from days to hours Content Moderation System Engineered a fine-tuned LLM-based content moderation pipeline achieving 95% classification accuracy on real-world data Automated flagging and routing workflows that eliminated 70% of manual review volume, directly reducing operational overhead Applied advanced prompt engineering and few-shot learning techniques to improve model output consistency across edge-case content categories Infrastructure & Quality Delivered a React.js analytics dashboard for real-time session monitoring, evaluation trends, and model performance tracking Automated the full regression test suite using Selenium, cutting regression testing effort by 30% and enabling faster sprint cycles

Education

Master of Technology, Artificial Intelligence, GH Raisoni University Amravati

Bachelor of Engineering, Computer Engineering, Sant Gadge Baba Amravati University, Amravati