Kenn Ejima, Developer in Tokyo, Japan
Kenn is available for hire
Hire Kenn

Kenn Ejima

Verified Expert  in Engineering

Full-stack Developer

Tokyo, Japan
Toptal Member Since
July 31, 2018

Kenn is an experienced full-stack developer who has built many web apps, API platforms, and native mobile apps from scratch over the past two decades. His core skills are versatile—data modeling, API design, maintainable code, beautiful UI/UX, TCP/IP networking, and optimizing DB performance. Kenn wrote one of the most referenced articles about the technical comparison between MySQL 8 and PostgreSQL 10.


Admit AI
Remix, PostgreSQL, Drizzle, Socket.IO, WebSockets, OpenAI, OpenAI GPT-4 API...
Dot Product AI
Remix, React, Node.js, TypeScript, Python 3, FastAPI, Vector Data, Qdrant...
Linux, Ruby on Rails (RoR), Ruby




Preferred Environment

FastAPI, Remix, PostgreSQL, Tailwind CSS, Ruby on Rails (RoR), TypeScript, Python 3, Qdrant, OpenAI, Cloudflare

The most amazing...

...thing I've built is a top-5 ranked mobile games' back end, which scaled from 0 to 30 million users in three years at under $3,000/mo to handle millions of DAUs.

Work Experience

Co-founder and CTO

2023 - PRESENT
Admit AI
  • Built a full-fledged LLM app where GPT-4 Turbo reviews college admissions essays in 10 days while working closely with a business domain expert through wireframe using Miro.
  • Used the JSON mode of GPT-4 Turbo and @streamparser/json to parse the streamed response from OpenAI and applied the proofread result and comments in real-time on the user interface using WebSockets/
  • Iterated over natural language processing to validate if the user submission looks like a real essay.
Technologies: Remix, PostgreSQL, Drizzle, Socket.IO, WebSockets, OpenAI, OpenAI GPT-4 API, Render, Sentry, Postmark, DMARC, Cloudflare, Docker, Node.js, Tailwind CSS, Radix UI, TypeScript


2023 - PRESENT
Dot Product AI
  • Developed the entirely new AI app with a brand new tech stack—Remix front end and FastAPI back end.
  • Incorporated OpenAI models to implement RAG (Retrieval Augmented Generation) using the Qdrant vector database.
  • Sent back a chat response with SSE (server-sent events) for a faster response.
Technologies: Remix, React, Node.js, TypeScript, Python 3, FastAPI, Vector Data, Qdrant, PostgreSQL, Cloudflare, Prisma, MailerSend, Tailwind CSS, Framer Motion, Socket.IO, Server Sent Events (SSE), OpenAI, Sentry, PostHog, Vitest, Retrieval-augmented Generation (RAG), Embeddings from Language Models (ELMo)


2017 - PRESENT
  • Provided support for MySQL, PostgreSQL, MongoDB, and Redis development.
  • Built an agent library for Rails that safely runs inside the customer's app.
  • Developed an AWS federation to generate a temporary authorization to S3.
  • Integrated a Stripe subscription with webhook handling.
  • Implemented a fast aggregation for the charts in the dashboard.
Technologies: Linux, Ruby on Rails (RoR), Ruby

Head of Japan

2018 - 2022
  • Contributed over the four years to the Japanese Quora growth: it was the fastest among different languages—a cross-functional team with the i18n product, community managers, moderators, and comms.
  • Built an automated framework with an ML model to predict the LTV of a cohort of users and bid per CPA accordingly in the US/English market—a cross-functional team with marketing, data science, and engineering.
  • Executed (Japanese i18n PM), from NLP to product design, a wide range of core product improvements for the Japanese based on split testing and persuasive intuition as one of the power users.
Technologies: Python, Jupyter Notebook, Google Sheets, Machine Learning, Machine Learning Operations (MLOps)

Co-founder and CTO

2013 - 2017
East Meet East
  • Managed a team of highly motivated developers and designers.
  • Implemented a multifaceted profile search and ranking algorithm with many fields, including proximity.
  • Developed antispam methods, including AI/machine learning assistance, image analysis, hell banning, and more.
  • Enabled subscriptions via iOS in-app purchase and Stripe.
  • Used a binomial proportion confidence interval on profile visits to approximate popularity rating.
Technologies: Swift, Redis, MySQL, Ruby on Rails (RoR), Ruby, NGINX, Linux, Objective-C

Chief Architect

2009 - 2013
Pankaku (Acquired by Colopl)
  • Built and scaled Pankia, a multitenant BaaS for iOS and Android, from zero to 30 million users in three years, where multiple iOS apps ranked in the top five in the US.
  • Designed a P2P protocol (semi-TCP over UDP) that passes through the firewall between mobile devices.
  • Kept the infrastructure cost under $4,000 with few back-end engineers.
  • Provided multi-language support for resources including leaderboards, achievements, items, and grades.
  • Composed a novel leaderboard algorithm where every user can know their ranked position with a good spatial-temporal tradeoff, even with millions of players using exponential buckets for scores.
  • Implemented Knuth's on-the-fly algorithm for variance applied to geometric mean to detect an anomaly in the response time data of HTTP endpoints.
Technologies: Objective-C, HAProxy, Redis, MongoDB, MySQL, Ruby on Rails (RoR), Ruby


2005 - 2009
Infoteria USA
  • Incorporated a US entity for a Japanese company which is now publicly traded.
  • Built Lingr, one of the world's first chat apps that used Comet for a persistent connection via HTTP.
  • Rented space in a data center in San Francisco, built servers, a Cisco router, and so on, and set them up on-site with the console.
Technologies: Java, MySQL, Ruby on Rails (RoR), Ruby


2000 - 2005
  • Identified requirements, determined specifications, and managed the release timelines for EAI/ETL products.
  • Managed product integrations at large enterprises in different industries in Japan—from manufacturing to finance to telecom to real estate.
  • Joined multiple committees to define XML-based industry standards such as RosettaNet, ebXML, etc.
Technologies: Java


1998 - 1999
Oracle Japan
  • Configured and performed benchmarks to beat competitors for Oracle OLAP products, using the highest-end hardware, such as the Sun Starfire series.
  • Consulted for Oracle's most important customers as a data warehousing specialist to define schemas for analytics.
  • Successfully improved the performance of a huge SQL with more than 100 lines of code.
Technologies: OLAP, Oracle

This project involved a database backup as a service. I built everything from scratch—including the back-end, the front-end, the API, the API consumer library, the Linux infrastructure, and the visual elements.

A key challenge was that we needed to support multiple databases such as MySQL, PostgreSQL, MongoDB, and Redis.

East Meet East

The number-one online dating app for Asian Americans. As co-founder and CTO, I built the infrastructure, the back-end, some of the front-end, and the early versions of the iOS app all the while managing the team of developers and designers.

As a consumer-oriented app, some of the challenges included fighting spam and fraudulent activities. Another unique problem with online dating is the multifaceted, multidimensional search for profiles and sort combined with multiple factors which included the Wilson score to represent desirability.


A BaaS for mobile game developers. It has all the features for a gaming platform that handles accounts, devices, friends, push notifications, leaderboards, achievements, and translations to multiple languages.

On top of that, we designed a fast-and-reliable P2P protocol that passes a firewall between devices which enabled game state sharing in the order of tens of milliseconds.
1994 - 1998

Bachelor of Engineering Degree in Applied Mathematics (Computer Science)

Kyoto University - Kyoto, Japan


Network Specialist (Japan)

Ministry of Economy, Trade, and Industry


Oracle Master Gold



Stripe API, jQuery CDN, Sidekiq, Resque, Jemalloc, Facebook API, jQuery, React, Node.js, Socket.IO, Drizzle


NGINX, Monit, Airbrake, Sentry, Git, Sublime Text, Google Sheets, Prisma, MailerSend


Ruby on Rails (RoR), Remix, Tailwind CSS


SQL, CSS, SCSS, Haml, Ruby, Objective-C, Java, JavaScript, Swift, Python, TypeScript, Python 3


iOS, MacOS, Kernel, Debian, Linux, Oracle, Jupyter Notebook, Docker


Redis, MySQL, Amazon S3 (AWS S3), Oracle RDBMS, PostgreSQL, MongoDB


Actor Model, OLAP


Comet, P2P, HAProxy, In-app Purchases, Machine Learning, Algorithms, Networking, FastAPI, Vector Data, Qdrant, Cloudflare, Framer Motion, Server Sent Events (SSE), OpenAI, PostHog, Vitest, Retrieval-augmented Generation (RAG), Machine Learning Operations (MLOps), WebSockets, OpenAI GPT-4 API, Render, Postmark, DMARC, Radix UI, Embeddings from Language Models (ELMo)

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.


Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring