Skip to content
Kevin Shah

Hi, I'm

Kevin Shah

Software Engineer | AI Engineer

I build and operate production LLM systems — agentic workflows, RAG pipelines, and multi-cloud AI platforms with systematic prompt evaluation, observability, and safety guardrails.

I take AI systems from prototype to production: LLM orchestration platforms, retrieval-augmented pipelines, prompt evaluation/optimization, and the observability and guardrails that keep them reliable at scale. Currently a Software Engineer on the Machine Learning Operations platform team at DNV; previously a Software Development Engineer at Amazon. Open to AI Engineer, MLOps, and Forward Deployed roles in the USA, Canada, Europe, and remote.

📍 Houston, TXOpen to USA, Canada, Europe & Remote

Experience

My professional journey building LLM systems, ML platforms, and backend services.

Software Engineer, Machine Learning Operations

Current
DNV·May 2023 – Present·Houston, TX

Design, build, and operate the AI platform behind agent-based products used across the business — retrieval pipelines, agent orchestration, evaluation, observability, and safety guardrails.

Grew through four phases in three years, partnering with product and engineering teams company-wide to turn use cases into shipped platform capability. Promoted March 2026.

AI Agent Platform

Apr 2026 – Present
  • Productionizing an internal agent platform for organization-wide use — agent definition versioning, deployment and rollback workflows, and operational support for agents embedded in customer-facing products.
  • Built agent evaluation infrastructure: offline and regression harnesses, task- and step-level scoring, golden datasets, and automated quality gates that catch behavioral regressions before release.
  • Implemented end-to-end tracing and observability — span-level instrumentation of agent reasoning and tool calls, token and latency metrics, cost attribution, dashboards, and alerting.
  • Added a model gateway layer providing request routing, provider fallback, rate limiting, and cost attribution across multiple model providers.
  • Extended the platform with file handling, reusable agent skills, a sandboxed code interpreter, and standards-based tool integrations connecting agents to internal data sources.
  • Led the front-end architecture and UX redesign covering agent authoring, trace visualization, and evaluation review.

LLM Platform Engineering

Mar 2025 – Apr 2026
  • Took ownership of a legacy LLM API service and re-architected it into a multi-provider platform serving teams across the organization.
  • Designed agentic workflow orchestration with persistent state, supporting multi-step reasoning and fault tolerance across long-running tasks.
  • Built retrieval-augmented generation pipelines end to end — document ingestion, chunking, embedding generation, vector indexing, and semantic retrieval.
  • Developed long-term agent memory services using top-k semantic retrieval to give agents durable, context-aware recall across sessions.
  • Built a prompt management system with LLM-assisted optimization and systematic evaluation — reduced hallucinations and cut a recurring 20-hour manual task to 3–4 hours.
  • Developed a knowledge-synthesis system that compiles cross-referenced documentation for model context injection, substantially reducing token usage and response latency versus full-context approaches.
  • Deployed inference-layer guardrails across environments via infrastructure-as-code — content safety, PII filtering, and hallucination mitigation.

Machine Learning Operations

Jul 2024 – Mar 2025
  • Built a platform-agnostic ML workflow execution engine pairing managed training with lightweight container workers — sub-5s execution for non-ML steps and ~40% faster processing overall.
  • Developed ML training pipelines that generate evaluation metrics and publish trained models to an internal data lake for downstream inference, including anomaly detection.
  • Re-architected a legacy data-lake execution service into a serverless design, cutting infrastructure cost by ~50% and lowering operational overhead.
  • Built and scaled REST APIs connecting ML models to internal products, working with QA and product to resolve integration issues.

Full-Stack Application Development

May 2023 – Jul 2024
  • Built a low-code website builder with custom drag-and-drop component blocks, letting users compose, style, preview, and publish pages from an internal component library.
  • Implemented page deployment pipelines on managed cloud infrastructure — CDN distribution, DNS, and infrastructure-as-code.
  • Designed and shipped a full-stack reporting tool that replaced a third-party product and eliminated its recurring license costs.
  • Extended reporting with SQL-based grouping, sorting, and filtering, and added functional test coverage for backend services.
PythonFastAPIAWSVector SearchInfrastructure as CodeCI/CDObservability

Software Development Engineer

Amazon.com Services LLC·Jun 2022 – Mar 2023·Austin, TX
  • Contributed analysis and design for a promotion-code migration spanning 21 global marketplaces — part of a zero-downtime effort that preserved the experience for 550M+ customers.
  • Researched multiple legacy codebases behind a 10+ year-old global Order Summary system and designed a change to its shipping section, including A/B test planning, requirements gathering, and integration analysis across several Tier-1 services.
  • Built runtime customer-experience monitoring to detect aberrations, supporting global latency tracking and real-time impact measurement.
  • Completed on-call ramp-up in two months against a three-month norm; served two rotations, each requiring diagnosis of latency and customer-impact issues and ownership of the operations-review handoff for a 17-member global team.
  • Authored test cases, participated in code review, and remediated existing service code to improve maintainability.
JavaPythonSQLShellAWS

Education

Master of Science in Computer Science

Arizona State University·GPA 3.81 / 4.00·May 2022·Tempe, AZ

Bachelor in Computer Engineering

Sardar Vallabhbhai National Institute of Technology (NIT Surat)·Jul 2020·Surat, India

Skills

Technologies and tools I work with day-to-day.

MLOps / AI

LangChainLangGraphMCPLiteLLMRAG PipelinesPrompt Engineering & OptimizationLLM EvalsAgent EvaluationRegression & Offline EvalsGolden DatasetsLLM Tracing & ObservabilityModel Gateways & RoutingAWS BedrockBedrock AgentCoreBedrock GuardrailsAzure OpenAIAzure AI FoundryGoogle Vertex AIGroqAgentic WorkflowsEmbeddings & Context Retrieval

Languages & Scripting

PythonJavaScriptTypeScriptSQLBash

Backend & API

FastAPINode.jsRESTful APIs

Cloud & DevOps

AWS LambdaAWS ECSAWS ECRS3SageMakerCloudWatchCloudFormationDockerGit

Databases

MongoDB (Vector)PostgreSQLRedis

Frontend & Reliability

Vue.jsNuxt.jsReactD3.jsUnit & Functional TestingProduction Observability

Get In Touch

Open to AI Engineer, MLOps, and Forward Deployed roles in the USA, Canada, Europe & remote.

Whether you have a role in mind, want to collaborate on a project, or just want to say hi — my inbox is always open. I'll do my best to get back to you promptly.

Email
kevinjshah2207 [at] gmail [dot] com

Location
Houston, TX · Open to USA, Canada, Europe & Remote

* Email shown as plain text to reduce spam. Use the form to send a message directly.