Shermayne Lee

Senior Full Stack Software Engineer, Generative AI & Agentic Systems

Perth, Western Australiashermayne.devLinkedIn

Profile

Full stack engineer with over nine years across React, Next.js, TypeScript, Node.js and Python. The part I care about most is the product itself. Right now I'm at a healthcare AI company building clinical AI products. A lot of that is agent orchestration, ASR evaluation and applied AI, but it all comes back to shipping something clinicians can actually use. Claude Code is a normal part of how I write code these days. Before this I spent seven years consulting across mining, oil and gas, and government, helping clients build products on the cloud and running AWS Well-Architected reviews. I'm looking for a senior role where I can keep doing product-focused engineering with some applied AI and client work in the mix.

Technical skills

Generative AI & Agentic Systems
AWS Bedrock, LangChain, custom agent harness design, multi-agent orchestration, working knowledge of MCP and A2A protocols, RAG, LLM-powered features, ASR evaluation
AI-Assisted Development
Claude Code (daily use for scaffolding, refactoring, test generation, code review), agentic development workflows
Frontend
React, Next.js, TypeScript, JavaScript, Vite, Tailwind CSS, design systems, SSR/SSG, state management
Backend & APIs
Node.js, Express.js, Python, REST, GraphQL, OAuth2, JWT, event-driven architecture, microservices
Cloud & Infrastructure
AWS (Lambda, API Gateway, S3, CloudFront, DynamoDB, ECS), Azure App Services & Functions, Terraform, CloudFormation, CDK, Docker, basic Kubernetes
Delivery & Quality
CI/CD (GitHub Actions, Bitbucket Pipelines), TDD, BDD, Jest, Cypress, Playwright, trunk-based development
Observability
OpenTelemetry, Grafana, Splunk, CloudWatch, Sentry
Data Engineering
Databricks, dbt, Fivetran, Spark, Snowflake, Python-based data workflows

Experience

Harrison.ai, Full Stack Engineer, Generative AI

Jan 2026 – Present · Perth, WA

  • Build AI products that help radiologists work more efficiently, cutting the time they spend on reporting and repetitive steps so they can focus on reading scans, working across the React and TypeScript front ends and the Go services behind them.
  • Design and maintain an in-house agent orchestration layer that coordinates multi-step LLM workflows, tool calls and state across cloud LLM platforms (including AWS Bedrock).
  • Evaluate our automatic speech recognition pipelines against clinical benchmarks, tuning prompt design and post-processing to bring down error rates so radiologists spend less time correcting their dictated reports.
  • Prototype orchestration patterns in LangChain alongside our own harness, so we can compare how quickly we can build against how much control we need for clinical work.
  • Use Claude Code as a standard part of the daily workflow for scaffolding, refactoring, test generation and code review, cutting time spent on repetitive implementation work.
  • Apply core networking and health data protocols (TCP, DNS, HTTP/TLS, MLLP, HL7/FHIR) to integrate agent services with clinical data pipelines carrying sensitive patient information.
  • Instrument services with OpenTelemetry for distributed tracing across the agent harness, and work with Kubernetes-based deployments for service orchestration.

Franklin.AI, Senior Software Engineer, Digital Pathology & AI Product

Aug 2025 – Dec 2025 · Perth, WA

  • Built scalable, data-rich interfaces in a multi-repo React/TypeScript monorepo supporting real-time image exploration for clinical workflows.
  • Developed reusable UI components and patterns aligned with internal design standards, cutting implementation effort across product teams.
  • Optimised rendering and interaction performance in large SPA environments handling complex, image-heavy data sets.
  • Integrated API-driven features with backend and product teams, with a focus on reliability and error handling across asynchronous workflows.
  • Refined shared packages, documentation and local development workflows to reduce friction and speed up iteration across the monorepo.

Mechanical Rock Pty Ltd, Senior Software Engineer & Consultant

2019 – Jul 2025 · Perth, WA

  • Led cloud-native delivery for enterprise clients across mining, oil and gas, and government, combining engineering practice with a product-focused approach.
  • Partnered directly with clients to define technical requirements and solution architecture, translating business problems into scalable delivery plans.
  • Conducted AWS Well-Architected Reviews, identifying security, reliability and cost gaps and delivering recommendations adopted by client engineering teams.
  • Designed internal developer tooling to streamline environment provisioning and automate secure deployment of pre-configured cloud infrastructure.
  • Provisioned cloud environments with Terraform and CloudFormation, ensuring consistency and auditability across multiple concurrent projects.
  • Built automated CI/CD pipelines and testing strategies (TDD, BDD, Cypress, Playwright) to improve release confidence and delivery speed.
  • Debugged and resolved production issues across TypeScript, Python and Go services, restoring stability in live client environments.
  • Architected serverless applications on AWS and Azure using Lambda, AppSync, DynamoDB and EventBridge.
  • Improved observability using Grafana and Splunk for real-time monitoring, logging and alerting.

Square18, Frontend Developer

2017 – 2018 · Perth, WA

  • Worked with UX/UI designers to implement responsive, accessible web applications using HTML, CSS and JavaScript.

Certifications

  • AWS Certified Solutions Architect – Professional
  • AWS Certified Developer – Associate
  • AWS Certified Cloud Practitioner
  • AWS Technical Professional
  • AWS Business Professional

Education

Diploma of Programming, South Metropolitan TAFE

2017 – 2019

Bachelor of Process Engineering, Universiti Putra Malaysia

2012 – 2016

Community engagement

AWS Community Day ANZ · Latency Conference · Perth Data Engineering Meetup · She Codes Australia

References available on request via LinkedIn.