Shreyas R.

Hi! I'm Shreyas — a Fullstack AI Engineer who builds intelligent, scalable web applications and ships AI-powered automations that actually work in production. I combine 5+ years of engineering experience across agentic AI, fullstack development, cloud infrastructure, and computer vision with an MBA from the University of Maryland (GPA 3.71/4.0). That means I don't just write code — I design solutions that align with your business logic, scale reliably, and deliver measurable ROI. What I deliver: AI Application Development & Integration - Conversational AI chatbots, recommendation systems, and NLP-based solutions using Llama-index, LangChain, and LangGraph - AI agent architectures for customer support, data processing, content generation, and decision-routing (CrewAI, LangGraph, mem0 for long-term memory) - GenAI integrations — LLM orchestration (RAG, custom prompts, GuardrailsAI), voice AI (LiveKit, Pipecat, Kokoro TTS, Whisper STT), and multi-modal systems - Custom AI model training, fine-tuning, and tailored solutions for specific business needs End-to-End Workflow Automation - Automated pipelines using n8n, LangChain, LangGraph, and custom scripts — from intake through execution to reporting - Voice-to-voice AI agents with long-term memory, RAG, and internet search (LiveKit, Llama-index, mem0, Tavily) Fullstack Web Development - Scalable web applications using Django and FastAPI - Responsive, interactive user interfaces with React.js - RESTful APIs, database design, ORM, and clean UI/UX Cloud Infrastructure & Deployment - Designing and deploying cloud-based solutions on AWS (EC2, S3, Lambda, RDS) - CI/CD pipelines for continuous integration and deployment - Docker, Kubernetes, Postgres data pipelines Computer Vision & Data Pipelines - PyTorch models (92% accuracy in production), image annotation pipelines, data-validation microservices Recent impact: - Reduced API time-to-first-token by 90% using streaming architecture (FastAPI, Postgres, Responses API) - Co-built a voice-to-voice AI therapist with long-term memory, RAG, and internet search (LangGraph, LiveKit, Llama-index, mem0, Tavily) - Delivered a computer vision algorithm at 92% accuracy with a full SaaS dashboard for solar farm inspection (PyTorch, Django, Nginx) - Built a 300% faster image-annotation pipeline for AI construction inspections at a Series B startup - Authored "The Ultimate Guide to Voice AI: From Basics to Pro with LiveKit Agents" (published on Amazon) Technical Expertise: - Fullstack Development: Django, FastAPI, React.js, HTML, JavaScript, SQL - Cloud Infrastructure: AWS (EC2, S3, Lambda, RDS), Docker, Kubernetes, Nginx - AI & Machine Learning: Llama-index, LangChain, LangGraph, CrewAI, PyTorch, Prompt Engineering, GRPO Fine-Tuning Voice AI: LiveKit, Pipecat, Kokoro TTS, Whisper STT, Silero VAD - Backend: RESTful APIs, Database Design, ORM, PostgreSQL, CI/CD, Selenium Observability & Ops: LangFuse, GuardrailsAI, mem0 Why hire me? Unlike pure engineers, I bring MBA-level business thinking to every project. I understand your cost structures, timelines, and ROI needs — so the solutions I build aren't just technically sound, they create measurable business value. I focus on delivering clean, maintainable code that scales with your business, and I collaborate closely with clients to ensure their vision becomes reality. Credentials that back it up: - AWS Certified Developer Associate - 10+ AI/LLM certifications (LangChain, LangGraph, RAG, GRPO Fine-Tuning, Red Teaming, Voice AI, CrewAI) - NUS BEng, University Scholars Program (top 3% of cohort) - Published author with hands-on open-source projects (1Student1School multi-agent tutor, Local Translator) - Won SGD $10K entrepreneurship grant; featured in Business Times, Today, Lianhe Zaobao If you're looking for a Fullstack AI Engineer who can turn your ideas into intelligent, scalable applications — let's connect! Message me with your project requirements and I'll map out an architecture within 24 hours.