AI Startups to Watch
This article consolidates three related articles into one.
Table of Contents
- Tackling Healthcare Challenges: Ryota Migitaka on the Lessons and Risks of Building in the AI Era
- The Future of AI-Driven Startups — The Keys to Efficiency and Innovation
- Agentic AI and the Entrepreneurial Future | TIMEWELL
Looking for AI training and consulting?
Learn about WARP training programs and consulting services in our materials.
Tackling Healthcare Challenges: Ryota Migitaka on the Lessons and Risks of Building in the AI Era
From pharmaceutical MR to entrepreneur building an SNS-powered recruitment platform for medical institutions — Ryota Migitaka's journey has been one of constant iteration. Multiple programs, multiple pivots, and eventually arriving at the business he runs today. He sat down to talk about the turning point that WARP represented, and to share his honest perspective on what it means to be a founder in the age of AI. His message: "Just get out into the field and keep going until someone pays you." Below is the story of a founder working to change the future of healthcare.
Profile:
Ryota Migitaka
CEO, MedX Inc.
Began his career as a medical representative (MR) at a pharmaceutical company in 2019. After working in sales in Shizuoka, he moved to headquarters and took on international drug development evaluation and research. He subsequently participated in Globis MBA, the CHANGE program, and the Ministry of Economy, Trade and Industry's SHIDO program, experiencing four business pivots along the way. In December 2024, he was selected for Nankai Electric Railway's "beyond the Border" — Japan's first secondment-to-entrepreneurship scheme — and in April 2025 founded a company providing SNS-based recruitment support services for medical institutions. #pharmacist #MBA
Contents:
- From pharmaceutical MR to entrepreneur — the healthcare challenges that sparked a new path
- Finding the answer through pivots — how WARP accelerated his business growth
- "Having something I want to do is enough" — the extraordinary support structure WARP provides
- Toward a future where healthcare workers smile — Migitaka's field-first vision
- Summary
From Pharmaceutical MR to Entrepreneur — The Healthcare Challenges That Sparked a New Path
— Could you start by telling us about your current work?
Migitaka: I run MedX Inc., which provides recruitment support services using social media. We primarily use Instagram and TikTok to help medical institutions recruit healthcare professionals.
I started my career as an MR at a pharmaceutical company in 2019. After two or three years of sales work in Shizuoka, I moved to the head office and worked on evaluating and researching overseas drug development pipelines. Eventually I left the company, and in April this year I launched an SNS-based recruitment business.
— You made a significant transition from pharma to entrepreneurship. What drove that?
Migitaka: When I was in sales, I honestly felt there wasn't much I was uniquely contributing. But I could see there were all kinds of problems in hospitals and medical institutions. At the time I was enrolled in the Globis MBA program — it wasn't really that I wanted to start a company, more that I thought it'd be great to personally solve the problems right in front of me.
The connection with Hamamoto goes back about four years, when I participated in CHANGE, the corporate challenger program. After that I got involved in the Ministry of Economy, Trade and Industry's SHIDO innovator development program, and those activities gradually led me toward starting a business.
Finding the Answer Through Pivots — How WARP Accelerated His Business Growth
— What brought you to WARP?
Migitaka: I saw the information in Hamamoto's Facebook post. At the time I didn't know much about AI and had no experience actually building anything. But I knew AI was going to be essential going forward — I thought it would be useful to have a bit more knowledge as a common language when talking with engineers.
WARP also ran in the evenings, so I figured it wouldn't put too much burden on the work I was already doing. And it was free. That's what made me sign up.
— What programs had you been involved in before, and how did your business idea evolve?
Migitaka: In CHANGE, I was working on a referral letter business — a service to digitize the paper referral letters that clinics send to hospitals. But I figured out why electronic health record companies weren't doing it: it wasn't profitable. So I pivoted.
In the SHIDO program, I worked on a service for pharmacists. I have a pharmacy degree myself and a pharmacist license, so I tried to build a knowledge-sharing tool for cancer drug information — something that would let pharmacies without specialists easily access expert knowledge. I actually built an MVP and got about ten pharmacists to use it, but they were too busy during the day to use it, and they were only logging in after 8 PM. I concluded it wasn't viable as a business.
After that I participated in MIRAI, a healthcare-focused program run by St. Marianna University School of Medicine, and HATSU, an entrepreneur support program in Kanagawa Prefecture.
— You've participated in a remarkable number of programs. How did you eventually arrive at your current business?
Migitaka: It was around the winter two years ago, during my time in MIRAI. In my conversations with hospital staff, I kept hearing that the fees paid to recruitment agencies were a real burden. The pain around nursing recruitment was especially loud. Even though pharmacists were in demand, the stronger ask was for nursing staff — and given the market size and the scale of the problem, I shifted to focus on nurses.
I should mention: I founded my company through Nankai Electric Railway's "beyond the Border" program — Japan's first secondment-to-entrepreneurship scheme. The model is: transfer from Tanabe Mitsubishi Pharma to Nankai Electric Railway, then almost immediately get to start your company. I was selected in the final screening in December last year, and I founded the company in April of this year.
"Having Something I Want to Do Is Enough" — The Extraordinary Support Structure WARP Provides
— Having participated in WARP, what did you find most valuable about it?
Migitaka: The support was genuinely exceptional. Engineers like Naito-san and Ando-san would answer questions casually over chat, or say "Want to jump on a call for thirty minutes?" and immediately make themselves available online.
These are people whose time is worth tens of thousands of yen per hour — and they'd generously teach you everything with a kind of volunteer spirit. Attending a coding bootcamp costs serious money, but with WARP you get that level of support for free.
I was trying to build a tool that would automatically generate TikTok scripts and evaluate them — honestly, the technical bar was too high and I didn't get to completion. But the knowledge and experience I gained along the way were significant.
— Who would you say WARP is most suited for?
Migitaka: I'd especially recommend it for people who want to build something with AI but don't have any engineer friends. You don't need to have a solid business idea yet. A lot of programs won't accept you unless you can make a compelling pitch — "Does this really have market potential? Can it become a unicorn?" — but WARP accepts you with something as simple as "I have something I want to do."
I was in the first cohort, so honestly there were parts of the engineering content I couldn't keep up with. Even though the lectures were great, carving out time for hands-on work was difficult. But even then, the best thing I got from it was making friends with younger engineers. We're already talking about building something together down the line.
On a practical note: WARP supports registration fees for company incorporation (note: for eligible applicants only, subject to prior screening). For anyone looking to start a company, that's a genuinely helpful provision.
— Were there also benefits in terms of knowledge?
Migitaka: Learning about tools like Cursor and v0 was huge. Without WARP, I wouldn't have known that Cursor was growing faster by market cap than Zoom. Having the experience of building an app with text input means that when you actually want to do something, the gap between knowing and not knowing is enormous.
My resistance to AI has disappeared — I've moved from "I have no idea" to something more like competence. I won't pretend it's immediately useful to me (laughs), but I do feel I've gained knowledge that's going to be essential in the future.
Toward a Future Where Healthcare Workers Smile — Migitaka's Field-First Vision
— Looking ahead, what do you want to take on as an entrepreneur?
Migitaka: My vision is "for healthcare workers to smile." Right now it's only recruitment support, but in the future I want to improve the working environment and create a world where people can work flexibly.
Think of something like Timee — a platform where you can earn money in your free time. There are plenty of nurses who want to work at the bedside, but understaffing persists. In the care sector, there's already a spot-entry service called KAITECH. I want to build something like that across the healthcare sector as a whole.
By combining that with social media, I want to become someone who can change the healthcare industry with fresh thinking that isn't bound by existing norms.
— A final message for people who are thinking about taking the leap.
Migitaka: I'm the type who participated in a lot of programs and still took two or three years before actually starting a company. If I'm being honest with my past self, the message is: just do it. Even if you're in a program, unless you're generating results in the field, it means nothing. Use that time to be with customers instead.
When you're in a program you gain peers, and in some sense it feels comfortable. But that "feeling of doing something" can actually get in the way. So please — take that one step out into the real arena.
"Just get out into the field and keep going until someone pays you." Stop being swayed by other people's opinions. Face your customers, and keep going until you can give them what they genuinely need.
Summary
Through this interview, it became clear that the path to entrepreneurship is never a straight line. Starting as a pharmaceutical MR, going through numerous programs and multiple pivots — the accumulated effort it took to arrive at his current business is tangible.
What stood out was the candid message: "Programs are important, but don't rely on them too much." Even while praising WARP highly as a place to build AI knowledge and expand his network, his most passionate point was ultimately the importance of "producing results in the field."
Ryota Migitaka — continuing his challenge to transform the healthcare industry with a new approach unconstrained by existing norms. It was a conversation that left us genuinely optimistic that people like him are going to change that industry.
For details on the WARP program, please reach out via the inquiry form.
The Future of AI-Driven Startups — The Keys to Efficiency and Innovation
In recent years, the rapid advance of artificial intelligence has accelerated its adoption across the business world. For startups in particular, AI has become a critical key to operational efficiency and driving new innovation. This article explores real-world AI use cases at startups and the impact that continued AI development is likely to bring.
Contents:
- Business automation and efficiency through AI
- Creating innovation with AI
- The future of AI development and its challenges
- Summary
Business Automation and Efficiency Through AI
For startups, making the most of limited resources is central to success. That's why AI-powered business automation and efficiency have become such a focus. Harrison Chase, CEO of LangChain — the open-source AI software development tool — has described how his own company uses AI in a variety of ways.
Specifically, he noted the following applications:
- Email assistant: AI automatically drafts replies to incoming emails
- Customer support bot: AI handles customer inquiries
- Marketing bot: AI automates parts of the marketing function
- SDR (Sales Development Representative) bot: AI handles lead research and initial outreach
These are tasks that were traditionally handled by junior hires or interns. By deploying AI, human resources can be redirected toward creative, higher-value work. However, Chase notes that because AI automation is not yet perfect at this stage, the "human in the loop" approach — where final judgment remains with people — remains important.
Creating Innovation with AI
Beyond automation, AI is expected to enable startups to generate new forms of innovation — developing products and services that were not previously possible.
In healthcare, for example, AI-powered diagnostic support systems and drug discovery platforms are advancing. In finance, AI-driven credit scoring and fraud detection systems are being deployed. These innovations represent things that were difficult or impossible to achieve with conventional methods.
Making AI-enabled innovation succeed requires clearing challenges that go well beyond the technical — building viable business models, navigating regulatory requirements, and more. Startups are expected to attack these challenges head-on and generate new forms of value.
The Future of AI Development and Its Challenges
AI technology will continue to develop, and its deployment in business will accelerate. Chase predicts that in the near future, AI agents will become significantly more capable — working more like human colleagues. Multiple AI agents collaborating to automate complex workflows will also become possible.
That said, there are real challenges. Questions about the transparency and accountability of AI-driven decisions, security of AI systems, and the impact of AI on employment are all recognized as serious social and ethical concerns. Navigating these challenges appropriately while extracting maximum value from AI will be critical.
At this point in time, building AI agents still requires significant technical expertise, making it difficult for organizations without specialized development teams. Going forward, more accessible AI development platforms are expected to emerge — making it possible for people without programming skills to leverage AI.
Summary
AI represents both a major opportunity and a set of real challenges for startups. The potential is enormous: efficiency gains from automation, development of innovative AI-powered services, and much more. At the same time, the social and ethical implications of AI deserve serious consideration.
Startups need to pursue AI adoption actively while simultaneously engaging honestly with the technical and social responsibilities it brings. That combination is what will allow them to maximize the benefits of AI and drive business growth and innovation. The challenge ahead for startups working to create new value through AI remains an exciting one to watch.
Reference: https://www.youtube.com/watch?v=_e6pgQ8yvqI
Agentic AI and the Entrepreneurial Future | TIMEWELL
This is Ryu from TIMEWELL. Today I want to talk about how generative AI is changing the future — the latest trends in a technology that is fundamentally reshaping the foundations of our lives and businesses. The evolution of generative AI that began in the latter half of the 2020s represents that transformation. As Elon Musk (founder of SpaceX and Tesla) put it: "AI will redefine the future of humanity." AI has become something that touches all of society. In line with that, generative AI has grown beyond simple information processing and text generation into a technology that genuinely complements human creativity and judgment. Among the most closely watched developments within generative AI is agentic AI. Agentic AI goes beyond simple task automation — it has the ability to drive work forward at the project level. The evolution of generative AI is making "building a unicorn company alone" shift from ideal to reality.
The Era of Building a Unicorn Alone — New Possibilities Opened by Generative AI
Looking back, founding a unicorn company (an unlisted company valued at over ¥100 billion) required significant resources and teams. But with the spread of generative AI — and in particular the emergence of agentic AI — the era of individuals running large-scale businesses has arrived.
The evolution of agentic AI has made "project-level work" possible, going beyond simple task processing. Anthropic's latest Claude AI model, for example, can autonomously operate a PC, coordinate across multiple applications, and drive an entire project forward.
Example: Claude's agentic capabilities — ChatGPT's agent functions can operate within a user's PC environment and automate sequences of tasks including email drafting, data analysis, and report generation.
Example: Salesforce's AI tools — Salesforce offers AI solutions that streamline sales activities and customer management, helping companies improve profitability.
These tools don't just allow "one person to handle multiple jobs" — they provide the capability for "one person to run an entire project."
Challenges in Generative AI Adoption: The Limits of Replit Agent and v0, and TIMEWELL's Solution
Among the most discussed generative AI app-building tools are Replit Agent and v0. These tools are groundbreaking in that they enable coding in natural language — meaning even users without programming skills can build applications. But this convenience comes with challenges.
Backend vulnerabilities: Applications built with Replit Agent or v0 tend toward weak backend design, precisely because they're so accessible. Data security and server load management are often insufficiently addressed.
Scalability limits: What works fine in the early stages can run into scalability problems as user counts and traffic grow. Sudden spikes in access can cause server outages and performance degradation.
Security holes: Relying on tool automation can introduce security vulnerabilities into the code, creating risks of data breaches or system compromise.
TIMEWELL's WARP Program: A Comprehensive Solution to Generative AI Adoption Challenges
TIMEWELL has developed WARP — a program designed to address exactly these challenges. It supports generative AI-powered application development with security and scalability built in from the start. Originally offered to a select group of companies, WARP has since been selected for Tokyo's SUTEAM startup entrepreneur support initiative, making it available to a broader range of aspiring and active entrepreneurs.
WARP program features:
- Security-first design: Provides best practices in API security and data encryption from the earliest stages of app development
- Scalable architecture support: Supports infrastructure design built for growing user counts and traffic, including load balancing and cloud environment optimization
- Prototype-to-production transition: Supports the move from prototypes built in Replit Agent or v0 to commercial-grade production systems
Generative AI Trends in China, the US, and Japan
Competition in generative AI is intensifying globally, with China and the US each leading through distinctly different approaches. China, in particular, is accelerating practical AI deployment through integrated ecosystems and government-backed policy support. The US, by contrast, emphasizes API-centric ecosystems that enable connections between independent applications.
China: Accelerating generative AI through integrated ecosystems
China's lead in generative AI is underpinned by integrated ecosystems — what might be called "giga apps." This model integrates multiple services within a single platform, delivering seamless user experiences.
WeChat (Weixin), operated by Tencent, is the defining example — a super app integrating messaging, payments, e-commerce, and lifestyle services. Tencent's generative AI-based chatbot automates customer service and content delivery. AI also handles automated generation of advertising copy and marketing messages, enabling efficient customer reach. Massive user behavior datasets are analyzed by AI to deliver personalized recommendations and improve engagement.
Alibaba uses generative AI to automatically create product descriptions and copy, helping sellers update their listings efficiently. Baidu's Ernie Bot — China's answer to ChatGPT — generates direct answers and content suggestions in response to search queries, transforming the conventional search experience.
The US: Building ecosystems through API connectivity
America's generative AI strategy focuses on API connectivity between independent applications.
OpenAI's ChatGPT API enables businesses to embed generative AI capabilities into their own services, with plugin functionality making it easy to connect external data sources and tools. Salesforce's Einstein GPT gives sales reps the ability to generate next-best-action recommendations based on customer data, draft emails, and produce presentation materials.
How the two approaches differ:
| Dimension | China | US |
|---|---|---|
| Integration model | Single-platform services integration (WeChat, Alibaba) | Independent services connected via API (OpenAI, Salesforce) |
| Leadership model | Government-led R&D support and regulatory coordination | Company-led innovation, diversity and competition |
Implications and challenges for Japan
Japan has much to learn from both China's integration efficiency and the flexibility and competitive dynamism of the US model. As more Japanese companies adopt generative AI, the key strategic priorities are:
- Building integrated ecosystems following the Chinese model, centralizing multiple services in unified applications
- Expanding API adoption following the US model, creating flexible API ecosystems to leverage existing generative AI tools
- Improving AI literacy — not just at the enterprise level but also enabling individuals to work with generative AI
The Skills That Will Be in Demand in the AI Era
The people who will be needed going forward are those who can "use generative AI to create new value." Specifically, the capabilities that will matter most are:
- Skills in using agentic AI to automate tasks
- The ability to understand market needs and propose solutions
- The technical competence to develop and operate applications end to end
TIMEWELL's Vision: Becoming an Event-Leading Company Powered by Generative AI
How will event management evolve over the next ten years? TIMEWELL aims to use generative AI and agentic AI to create entirely new forms of event operation — ones that break from the conventional playbook.
TIMEWELL has accumulated experience and know-how through more than 1,000 event operations, providing solutions that balance operational efficiency with quality. To advance that experience to the next stage, we are focusing on the evolution from generative AI implementation within applications to agentic AI.
The future of event management through agentic AI
Event management is a complex process spanning planning through execution to follow-up. TIMEWELL will use agentic AI not just to streamline operations, but to provide a next-generation event platform focused on what events are truly about: bringing people together and creating memorable experiences.
What we're aiming for is events that are "more fun" and "spaces that generate new value." Here's how agentic AI will reshape the event experience:
1. Support in the planning phase
Idea generation — proposing event concepts that create impact: Using generative AI to suggest unique event themes and content based on participants' interests and preferences.
Example: Quickly proposing "collaboration workshops" for attendees who share interest tags (sustainability, sports, art, etc.). Example: Gamified session structures aligned with specific themes that keep participant engagement high.
Schedule optimization: AI checks speaker and venue availability in real time and automatically suggests optimal schedules, reducing the operational burden while delivering value to participants.
Audience development: CRM integration with generative AI sorts target customers and automatically generates and sends optimized invitations and advertising, using past event participation data to deliver personalized recommendations.
2. Support in the operations phase
Day-of operations assist: Agentic AI automates check-in, seat assignment, and issue resolution to support event operations. Through integration with Brother Corporation, digital business card creation and name badge printing become possible on the spot.
Digital business cards and SNS features: Participants create and share digital business cards within the event to expand their networks, with social features that make it easy to maintain connections afterward.
Live feedback analysis: Sensor data and social media reactions are used to analyze participant responses in real time, with improvement suggestions that increase the energy of the event.
3. Utilization in post-event follow-up
Personalized follow-up: After the event, AI automatically generates follow-up emails and surveys based on participant behavior data. Matching based on shared interest tags facilitates new collaboration and business opportunities.
Example: Attendees who shared the "startup" tag reconnect online after the event and co-found a new venture.
4. Event content proposals that maximize fun and impact
Agentic AI proposes gamified content to enhance the "fun" and "impact" dimensions of events.
Example — gamification: Point accumulation systems and mini-contests with prizes create a sense of unity across the whole venue.
Example — storytelling that creates impact: Visuals and music are generated in real time in sync with speakers' remarks, maximizing the emotional impact on the audience.
TIMEWELL's Vision for 10 Years From Now: Events as the Starting Point for Connection and Value Creation
In ten years, TIMEWELL will have evolved from optimizing event operations into a leading company that provides a platform centered on "bringing people together" and "creating new value." The agentic AI-centric platform will deliver the following:
- Building a global event network: Agentic AI facilitates connections across regions and industries, sharing best practices and driving new collaboration
- Sustainable and inclusive operations: Paperless workflows through digital badges and online participation, making events more environmentally friendly and accessible to all
- Delivering fun and memorable experiences: Building systems where the emotional impact of events leads to subsequent collaboration and new value creation — AI-powered content proposals and matching that make participants say "I want to come back"
From Impact to the Future — Create New Event Experiences with TIMEWELL
TIMEWELL believes in the potential that emerges from people meeting one another. The future we're building isn't simply about efficiency — it's about creating event experiences defined by "impact" and "joy." We invite you to discover what AI can open up at your next event. The door to the future opens with us.
TIMEWELL will continue pursuing the possibilities of generative AI and agentic AI in event management, looking ahead to the next ten years. We commit to remaining a leader that innovates the future of events — never satisfied with past success.
The future of event management is evolving toward something more efficient, more personalized, and more inclusive. Will you begin this journey with us? TIMEWELL is fully committed to supporting your success. Let's open the door to the future together.
Streamline Event Management with AI | TIMEWELL Base
Struggling with the complexity of running large-scale events?
TIMEWELL Base is an AI-powered event management platform.
Track Record
- Adventure World: Managed Dream Day with 4,272 participants
- TechGALA 2026: Centrally managed 110 side events
Key Features
| Feature | Impact |
|---|---|
| AI page generation | Event page ready in 30 seconds |
| Low-cost payments | 4.8% processing fee (among the lowest in the industry) |
| Community features | 65% of attendees continue engaging after events |
For a conversation about streamlining your event operations, feel free to reach out.