Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am Stephen Ranshous, a Principal Full-Stack and AI Platform Engineer with 10+ years of experience building cloud-native, enterprise systems. I specialize in creating scalable, secure software architectures that deliver measurable business impact. I focus on LLM-powered platforms using RAG, agent workflows, and microservices, with a strong emphasis on security, explainability, and compliance. I excel at modernizing legacy systems and delivering AI solutions that drive tangible business value across large-scale deployments.…I am Stephen Ranshous, a Principal Full-Stack and AI Platform Engineer with 10+ years of experience building cloud-native, enterprise systems. I specialize in creating scalable, secure software architectures that deliver measurable business impact. I focus on LLM-powered platforms using RAG, agent workflows, and microservices, with a strong emphasis on security, explainability, and compliance. I excel at modernizing legacy systems and delivering AI solutions that drive tangible business value across large-scale deployments.

Stephen Ranshous

Full Stack Developer, AI Engineer, AI Developer, +3





I am Stephen Ranshous, a Principal Full-Stack and AI Platform Engineer with 10+ years of experience building cloud-native, enterprise systems. I specialize in creating scalable, secure software architectures that deliver measurable business impact. I focus on LLM-powered platforms using RAG, agent workflows, and microservices, with a strong emphasis on security, explainability, and compliance. I excel at modernizing legacy systems and delivering AI solutions that drive tangible business value across large-scale deployments.…I am Stephen Ranshous, a Principal Full-Stack and AI Platform Engineer with 10+ years of experience building cloud-native, enterprise systems. I specialize in creating scalable, secure software architectures that deliver measurable business impact. I focus on LLM-powered platforms using RAG, agent workflows, and microservices, with a strong emphasis on security, explainability, and compliance. I excel at modernizing legacy systems and delivering AI solutions that drive tangible business value across large-scale deployments.

Available to hire

I am Stephen Ranshous, a Principal Full-Stack and AI Platform Engineer with 10+ years of experience building cloud-native, enterprise systems. I specialize in creating scalable, secure software architectures that deliver measurable business impact.

I focus on LLM-powered platforms using RAG, agent workflows, and microservices, with a strong emphasis on security, explainability, and compliance. I excel at modernizing legacy systems and delivering AI solutions that drive tangible business value across large-scale deployments.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

English

Fluent

Work Experience

Principal Engineer (AI Focus) at Saviynt

January 1, 2023 - Present

Architected an agentic AI platform supporting AI-augmented full-stack features for an enterprise Identity Governance & Administration platform, improving access decision accuracy and audit readiness by 30%+. Orchestrated LLM + RAG-based workflows to automate identity certification, access reviews, and policy remediation, grounding AI decisions in governance policies. Built RAG pipelines indexing 500K+ identity records for context-aware, explainable access recommendations; enabled a RAG-enabled IAM chatbot improving answer accuracy by 30% and reducing manual policy lookup by 45%. Refactored legacy services into API-first microservices to enable 40% automation of access governance workflows and integrated agent workflows via LangGraph.

Senior Full-Stack Engineer at Cisco

January 1, 2019 - January 1, 2023

Architected agentic full-stack systems combining Java/Python backend services, event-driven pipelines, and React dashboards to help network operators detect, investigate, and remediate infrastructure issues across 1M+ devices, processing 500K+ operational events/day. Designed agentic architectures with AI/ML inference, policy-based rules, and recommendations, while maintaining human-in-the-loop controls for reliability and auditability. Implemented Python-based orchestration and automation services acting as control planes across 20+ distributed services to reduce manual intervention by 20%. Built operator-facing React + TypeScript applications to visualize system state and workflow progress, reducing MTTR in high-volume environments.

Full-Stack Engineer at ServiceNow

January 1, 2015 - January 1, 2019

Led innovation of a large-scale monolithic platform by introducing service-oriented boundaries and asynchronous, event-driven processing for long-running workflows. Extracted shared logic into reusable backend services, reducing latency and enabling multi-tenant SaaS delivery. Implemented Java back-end services and workflow engines powering the ITSM platform for thousands of enterprise customers, and tuned queries, background jobs, and front-end data loading to achieve ~20% faster workflow execution. Created Python automation scripts and data-processing utilities; developed React + TypeScript operator-facing apps to monitor workflows and performance.

Software Developer at BMC Software

January 1, 2013 - January 1, 2015

Programmed Java-based ITSM modules with state-driven workflows, escalation rules, and SLA tracking, handling 10K+ incidents monthly in large customer environments. Improved reliability of enterprise job scheduling and batch automation by implementing failure detection, retry mechanisms, and dependency resolution, reducing missed or failed executions by 25%. Developed REST and SOAP APIs enabling integration across 5–10 external systems, improving data synchronization reliability. Optimized Java backend services supporting monitoring and alerting pipelines, processing 100K+ events/day with emphasis on performance and fault tolerance.