Srivardhan Muthyala

AI Engineer with 3+ years of hands-on experience designing and deploying scalable Machine Learning workflows, robust Retrieval-Augmented Generation (RAG) pipelines, and intelligent conversational agents. Proficient in integrating industry-leading multi-cloud services (AWS, GCP, Azure), optimizing LLM prompt engineering, and managing containerized ML deployments. Demonstrated ability to collaborate with cross-functional teams to build data-driven applications that enhance user experience, reduce inference latencies by ~40%, and achieve substantial operational cost reductions.

Skilled in translating technical requirements into performant backend services with strong adherence to MLOps and CI/CD best practices.

Freelance Consultant - Master Pro Dev - Toronto, ON

(2025-09)

Project 1: Enterprise Multi-Agent Automation Platform | Client: Spinning Forward (Emerging Journalist and Content Creation Startup funded by Google)

• Architected enterprise-scale multi-agent AI systems using LangGraph, MCP, AutoGen, and tool-calling frameworks to automate support, lead qualification, analytics, and operations.
• Designed autonomous reasoning, retrieval, and orchestration agents capable of multi-step planning, contextual reasoning, and dynamic API execution.
• Built scalable RAG and inference infrastructure using Pinecone, PGVector, Kubernetes, Redis, vLLM, Prometheus, and Grafana for production-grade observability.

Senior AI Engineer - Master Builder Mercantile - Calgary, AB

(2025-05 - 2025-07)

Project: Enterprise Agentic AI Platform for Customer Support, Enterprise Knowledge & Product Automation. Business Requirement: Modernize customer engagement and internal operations by developing a unified enterprise AI platform capable of automating multilingual customer support, empowering employees with enterprise knowledge, and streamlining vendor product onboarding across digital commerce channels.

• Architected a secure, cloud-native Agentic AI platform on AWS, enabling three integrated business capabilities: multilingual customer support, an enterprise knowledge copilot, and automated product catalog management while maintaining isolated workloads through multi-tenant architecture and shared enterprise knowledge services.
• Engineered multilingual, multimodal, and multichannel AI assistants using LangChain, LangGraph, and MCP, implementing enterprise RAG, tool calling, conversational workflows, semantic and episodic memory, vendor catalog ingestion, schema normalization, AI-assisted SEO generation, and Shopify publishing automation.
• Established production-grade LLMOps, AgentOps, and RAGOps by implementing secure CI/CD, observability, role-based access control, automated evaluation pipelines, audit logging, retrieval optimization, workflow monitoring, and enterprise governance across production AI workloads.
• Automated customer support operations, accelerated employee knowledge discovery, standardized vendor catalog publishing, reduced manual processing effort, and improved enterprise AI adoption through scalable cloud-native architecture and production-ready operational practices.

Senior AI Engineer - OZ Capital - Concord, ON

(2024-09 - 2025-05)

Project 1: Enterprise Conversational AI Platform. Business Requirement: Modernize enterprise customer engagement by developing a multilingual conversational AI platform capable of automating voice and digital interactions, retrieving enterprise knowledge, and integrating seamlessly with backend business applications to improve customer experience and operational efficiency.

• Architected a cloud-native conversational AI platform supporting multilingual voice and chat interactions, integrating enterprise knowledge retrieval, intelligent intent routing, contextual conversations, secure API integrations, and human-agent escalation for complex customer support scenarios.
• Developed conversational workflows using LangChain, enterprise RAG, prompt orchestration, conversation state management, Redis-backed session caching, semantic conversation history, and contextual query rewriting to deliver personalized, multi-turn customer interactions.
• Engineered scalable backend services using FastAPI, containerized microservices, REST APIs, automated CI/CD pipelines, cloud-native deployments, and secure integrations with enterprise CRM, email, and ticketing systems to support reliable production workloads.
• Established production LLMOps practices by implementing centralized monitoring, prompt evaluation, conversation analytics, automated testing, release validation, and operational observability while tracking intent accuracy, conversation completion, escalation rates, latency, cache utilization, and user satisfaction.

Senior AI Engineer - OZ Capital - Concord, ON

(2024-09 - 2025-05)

Project 2: Cross-Cloud Enterprise Knowledge Platform. Business Requirement: Build a secure cross-cloud Retrieval-Augmented Generation (RAG) platform enabling enterprise users to retrieve contextual knowledge from distributed AWS and Azure data sources while ensuring low-latency retrieval, governance, and high-quality AI-generated responses.

• Architected a cloud-native enterprise RAG platform integrating distributed knowledge repositories across AWS and Azure, enabling secure document ingestion, semantic search, contextual retrieval, and intelligent response generation for enterprise knowledge management.
• Engineered scalable document ingestion and indexing pipelines incorporating OCR, metadata enrichment, document parsing, semantic chunking, embedding generation, vector indexing, hybrid retrieval, reranking, and contextual compression to improve retrieval accuracy and response relevance.
• Implemented secure multi-tenant retrieval pipelines using LangChain, conversation-aware retrieval, metadata filtering, Redis-backed retrieval caching, embedding lifecycle management, and automated document synchronization to support scalable enterprise knowledge access.
• Established production RAGOps practices by implementing automated indexing workflows, retrieval evaluation, infrastructure observability, CI/CD automation, and continuous quality monitoring using RAGAS metrics including Faithfulness, Answer Relevance, Context Precision, Context Recall, hallucination rate, retrieval latency, and citation quality.

Senior AI Engineer - OZ Capital - Concord, ON

(2024-09 - 2025-05)

Project 3: Enterprise LLMOps & Multi-Cloud AI Platform. Business Requirement: Standardize the lifecycle of enterprise Large Language Models by building a cloud-native LLMOps platform for experimentation, evaluation, deployment, and production monitoring across AWS, Azure, and GCP, ensuring scalable, governed, and reliable AI delivery.

• Designed enterprise LLMOps workflows using Amazon Bedrock, Amazon SageMaker, PEFT, LoRA, and MLflow to streamline model experimentation, fine-tuning, prompt optimization, model versioning, deployment, and lifecycle governance for multiple Generative AI applications.
• Architected a standardized multi-cloud AI serving platform using Azure Machine Learning, Vertex AI, Docker, Kubernetes, Helm, and automated CI/CD pipelines, enabling consistent deployment, scaling, rollback, and operational governance across cloud environments.
• Established production AI evaluation and observability pipelines by combining LLM-as-a-Judge, BLEU, and ROUGE metrics for comprehensive model performance tracking.