
Rin Arakaki, Architect
Profile
Strategic and innovative Lead Software Engineer and Architect with over 9 years of experience designing, building, and leading implementation of scalable, resilient cloud-native solutions on AWS and GCP. Proven ability to define technology strategy, drive architectural decisions, lead high-performance teams, and align technical capabilities with business objectives.
Experience
Full-Stack Engineer, JAPAN AI
Oct 2024 — Current, Tokyo, Japan
LLM/generative AI research and development.
- Architected and led development of AI call centre platform, defining infrastructure architecture, database schema, UI, and AI agent design. Researched UDP-based stateful communication protocols and Kubernetes network model, integrated with third-party telephony provider. Presented demo to CEO.
- Implemented call centre operator and customer matching, using mathematical optimisation.
Technologies — ChatGPT, Gemini, Kubernetes, Google Kubernetes Engine, Cloud VPN, Cloud Interconnect, Cloud Batch, Cloud Run, Cloud Scheduler, Cloud Tasks, Cloud Storage, SIP, WebRTC, PostgreSQL, C++, Python, TypeScript, React/Next.js, Express, gRPC
Full-Stack Engineer, Kakeai
Sep 2023 — Sep 2024, Tokyo, Japan
One-on-one platform with real-time AI transcription and summarisation features.
- Re-architected and rebuilt core calendar component, with external calendar syncing, recurrence rules, third-person event creation, and preferred time slot setting. Reduced complexity and codebase size by two-thirds and improved maintainability.
- Proposed fine-grained access control model for multi-tenant SaaS architecture to CTO.
- Analysed engagement data using causal inference (Python).
- Led mentor-mentee matching implementation using stable matching algorithm.
Technologies — Amazon Elastic Container Service, Terraform, Python, Ruby on Rails, TypeScript, Angular
Lead Back-End Engineer, Viagate
Apr 2023 — Aug 2023, Tokyo, Japan
Marketing research platform using eye tracking, and facial expression recognition.
- Designed, built scalable, cost-effective AWS data infrastructure from scratch; employed binary serialisation, event-driven architecture for time-series data ingestion, aggregation, compression, querying.
- Designed database schemas, APIs, storage hierarchies; defined clear, consistent, broadly applicable interfaces across data infrastructure. Introduced Protocol Buffers (iOS, web front-end) ensuring schema consistency, minimising time-series data volume.
- Led backend team developing APIs fully leveraging PostgreSQL functionalities.
- Implemented video, image processing (Python, OpenCV) for data visualisation.
Technologies — AWS IAM, Amazon Cognito, AWS AppSync, Amazon EventBridge, Amazon Aurora PostgreSQL, Amazon S3, AWS Lambda, Python, Apache Arrow/Parquet, Protocol Buffers, GraphQL
Technical Lead, Delipicks
Jun 2019 — May 2020, Tokyo, Japan
B2B food delivery startup; joined 6 months after launch.
- Served as Technical Lead, defining technology strategy and aligning system development with business processes through Lean methodologies. Achieved estimated cost reductions equivalent to 12 customer support staff salaries annually.
- Identified future technical challenge/opportunity critical to viability and scalability of business model. Conducted technical research and proposed innovative solutions to CEO, using mathematical optimisation and genetic algorithms.
- Actively participated in full business development lifecycle: cold calling, establishing supplier relationships, deploying operations app to partner workplaces, and negotiating deliverables.
- Participated in UX workshops, implemented UI prototypes and conducted user interviews.
- Managed full recruitment cycle (sourcing, interviewing, hiring, onboarding) for two engineers, one mathematical optimisation specialist.
- Led a cross-functional team of five, including two software engineers, one QA, one designer and one project manager.
- Managed real-time coordination with customers, suppliers, delivery partners to resolve operational and system issues.
Technologies — Firebase, Python, JavaScript, React
Software Engineer, Styler
Aug 2018 — Mar 2019, Tokyo, Japan
Fashion retail platform connecting online interactions and in-store experience (OMO - Online-Merge-Offline / New Retail).
- Led technical decisions for infrastructure, database design, API development. Initiated and guided the architectural transition from a monolithic system to a modern SPA + API architecture, improving deployment frequency from from bi-weekly to weekly.
- Built new engineering team from scratch as part of a team of 6 engineers. Participated in transition from offshore outsourced to in-house development.
- Developed categories and tags databases for e-commerce platform, enhancing both on-site search functionality (linked from owned media) and SEO performance.
Technologies — Google Kubernetes Engine, Cloud Functions, Cloud SQL, Ruby on Rails, Python
Software Engineer, MarketEnterprise
Apr 2016 — Dec 2017, Tokyo, Japan
Resale and rental business with complex operations.
- Architected and led development of organization-wide authorisation API, initiating legacy system modernisation into microservices architecture; synthesised CTO's project vision, conducted research, made key design decisions. Reduced API response latency from 7s in experiment to 0.2s in production.
- Implemented social sign-in functionality for e-commerce website, managing development from front-end and server-side logic through to database migration.
- Led data aggregation and transformation efforts deriving user behaviour insights. Collaborated closely with a public relations and a fulfilment manager.
Technologies — Amazon EC2, Amazon RDS MySQL, Amazon DynamoDB, Amazon API Gateway, AWS Lambda, Python, JavaScript
Education
The Open University of Japan, Bachelor's Degree, Liberal Arts
Apr 2012 — Mar 2016, Tokyo, Japan
- Completed degree requirements in 2 years.
Projects
Founder, Synthetic
Dec 2020 — Present, Tokyo, Japan
AI-powered formal mathematical proof platform.
- Researching, developing database and graph/network-based search system for mathematical formulae.
- Investigating theorem provers and proof search engines.
- Researching formal proof generation based on LLM and reinforcement learning.
Technologies — Rust, Python, TypeScript, React/Next.js, Express