Headline

AI Engineer at Protean

About

I build production-grade AI systems that solve real business problems at scale. Over the past 3+ years, I've architected enterprise AI solutions — from custom RAG frameworks and AI chatbots to OCR pipelines and ML automation systems — that have directly impacted millions of users. Here's what I've shipped: -> Reduced cloud infrastructure costs by 60% by building a custom RAG framework with open-source LLMs and vector databases, replacing expensive vendor solutions while improving response quality by 40%. -> Architected an Agentic AI chatbot using Vertex AI, Autogen, and LangChain — achieving 90% intent accuracy and 90% user satisfaction across 10,000+ daily interactions. -> Deployed an OCR pipeline processing 30,000+ PAN documents daily with 95%+ accuracy, with GPU-accelerated preprocessing that improved throughput by 3x. -> Automated email response workflows using ML classification models, processing 500+ emails daily and reducing turnaround time by 70%. -> Built an LLM observability platform with OpenTelemetry and Grafana for real-time monitoring of model performance, drift detection, and cost tracking. My technical toolkit: - ML/DL: PyTorch, TensorFlow, Scikit-learn, XGBoost, Transformers - LLM/AI: LangChain, RAG, Autogen, Hugging Face, LoRA/QLoRA, Prompt Engineering - NLP: Text Classification, NER, Sentiment Analysis, Semantic Search - MLOps: Docker, Kubernetes, MLflow, CI/CD, Model Monitoring - Cloud: AWS (SageMaker), GCP (Vertex AI), Azure ML - Data: PostgreSQL, MongoDB, Qdrant, FAISS, Pandas, NumPy I'm passionate about building AI systems that are not just technically sound but actually work in production — handling real traffic, real edge cases, and real business constraints. Currently exploring opportunities in ML Engineering, AI Engineering, and Applied AI roles where I can build and scale impactful AI products. Let's connect — always happy to talk about RAG architectures, LLM optimization, or anything AI. ayushmaheshwari068@gmail.com

Location

Pune District

AI stack

anthropic apis

antigravity

claude code

claude

codex

cursor

Work experience

AI Engineer at Protean eGov Technologies Ltd. (Formerly NSDL e-Governance Infrastructure Ltd.), Mar 2023 - Present · 3 yrs 5 mos, Pune, Maharashtra, India, Building production-grade AI systems for enterprise e-governance platforms serving millions of users. Key Contributions: • Architected an enterprise-grade Agentic AI chatbot using Google Conversational Agent, Vertex AI, and Autogen — achieving 90% intent accuracy and 90% user satisfaction across 10,000+ daily interactions. • Developed a custom RAG framework using open-source LLMs and Qdrant vector database — reducing cloud costs by 60% and improving response quality by 40% over the previous vendor solution. • Automated email response workflows using TensorFlow and Scikit-learn — increasing operational efficiency by 45% and reducing turnaround time by 70%. • Built an LLM observability framework with OpenTelemetry — enabling real-time performance monitoring and drift detection across production models. • Designed scalable RESTful APIs in Django supporting 10,000+ daily requests for analytics and reporting workflows. • Implemented vector search pipelines with embedding models and Qdrant for semantic document retrieval, powering enterprise knowledge management systems. • Optimized end-to-end ML pipelines — from data preprocessing and feature engineering to model training, evaluation, and deployment — reducing model iteration cycles by 30%.

Programming Analyst Trainee at Cognizant, May 2022 - Feb 2023 · 10 mos, India, Full-stack engineering training with focus on enterprise application development, data visualization, and cloud systems. • Built interactive data visualization dashboards using Angular and JavaScript — improving BI reporting efficiency by 30% for analytics stakeholders. • Developed and deployed data processing pipelines for ETL workflows, handling structured and unstructured data from multiple enterprise sources. • Completed intensive training in microservices architecture, enterprise application development, and cloud systems (Azure). • Delivered multiple Agile-based project modules with 99% uptime and on-time delivery across 3 sprint cycles.

Data Science & Business Analytics Intern at The Spark Foundation at The Sparks Foundation, Aug 2021 - Sep 2021 · 2 mos

Education

Bachelor of Technology - BTech, Computer Science & Engineering (AI & ML), Galgotias university

Public links

github: https://github.com/Ayush0877

leetcode: https://leetcode.com/CA3fVfPOCt