Jinba (YC W26)

Took the IVS Kyoto stage after Japan's former Prime Minister, on the Transpose Platform panel. Good conversation with Alex Bangash moderating, alongside Hideya Sadanaga, Ilana Elbaz, Panos Madamopoulos-Moraris, and William Salcedo. We traded ideas on what it actually takes to bridge the US and Japan markets. Japan's startup scene is moving. Good to be in the room.

We have crossed 100 pre-built integrations in Jinba :) OpenAI, Salesforce, Slack, HubSpot, GitHub, Gmail and many more. Your ops manager writes a prompt describing what they need. Jinba generates the workflow. It connects to the tools you already run on. You deploy it as an MCP, API endpoint, webhook, Slack bot, or cron job. We built Jinba for teams in banking and insurance, industries where automation is urgent but the tools have always asked too much of the people trying to use them. If your team has been sitting on automation ideas because the build work felt too heavy, this is worth a look.

co-hosting a few fintech / AI dinners in NYC and Boston in last week of May and first week of June. looking for 2-3 paid volunteers in each city to help us out for 4 hours with guest registration and co-running the event with us :) the guests are bank execs and senior technical staff so a great networking opportunity if you’re doing postgrad or are early in your tech career and wanna 10x your network in 1 night. dm me ✌️

 a16z Tech Week is coming to Boston, and then shortly after to New York City. We're Jinba (YC W26), building the AI layer for the world's largest banks, and we'd love to organize an event or two in both cities during the tech weeks. If you're a VC or startup or bank in Boston or New York City and would like to co-organize these events with us, please reach out to me ✌ Also, if you're a student or technology enthusiast and you are interested in helping us run the events, get to meet some of the brightest people in Boston and NYC, and get some cool merch, plus allowance for your help, please reach out to me too :)

The banks that will lead on operational efficiency over the next five years are not the ones with the largest technology budgets. They are the ones that stopped waiting for the perfect enterprise platform and started automating the workflows that are already fully defined. Every major bank has dozens of them. KYC document review. AML alert triage. Loan underwriting document routing. Contract clause extraction. Investment document assessment. Regulatory reporting preparation. The logic for every one of these workflows already exists. It is written down in compliance policy documents, operations manuals, and audit frameworks. The steps are defined. The decision rules are clear. The output requirements are known. The only thing missing is a tool built to run those workflows automatically, on-premise, with a full audit trail, in a way that both your compliance team and your regulator can inspect. That is what Jinba is built for. Not AI that replaces your compliance team. Not a six-month consulting engagement. Not a cloud platform that fails the security review. Deterministic workflow automation for the processes that govern how a Tier 1 bank actually operates. Deployed in days. Full audit logging at every step. Workflow logic your operations lead can read and verify without filing a ticket to engineering. The workflows are ready. The logic is already defined. The only question is how long you continue running them manually. Visit jinba.io or reach out directly to start the conversation.

🚀 Happening This Wednesday – SSS Growth Hub 🚀 The Growth Hub is our monthly gathering for open, honest conversations about what it really takes to build and scale ventures in Japan. This session, we dive into the tactical bridge between Tokyo and Silicon Valley — through the lens of Jinba (YC W26) (they're hiring!). Making it into Y Combinator is a statistical rarity. Jinba is one of only a handful of Japanese startups to achieve this in the past two decades. We’ll hear from Shoya Matsumori, a veteran of Tokyo’s ML ecosystem, who secured a spot after six consecutive rejections — and a bold move to San Francisco. Since launching in April 2025, Jinba has rapidly scaled its user base, validating the demand for natural language interfaces in complex operational environments. 💡 In this session, we’ll explore: ✅ What it actually takes to get into Y Combinator ✅ The reality of building from Japan and scaling globally ✅ Lessons from the SF move and Demo Day journey ✅ What Japan’s startup ecosystem can learn and apply 🎤 Moderated by Francisco Dalla Rosa Soares (Special Advisor, Shibuya City) 🗣️ Conducted in English 📍 Shibuya Startup Support 📅 Wednesday, April 22 4:30 PM - 6:00 PM 🎫 Register here 👉 https://luma.com/h89ehg91 💼 Jinba is hiring — and actively looking for technical talent. Don’t miss this opportunity to connect directly. Shoya Matsumori Jinba (YC W26) Francisco Dalla Rosa Soares Yaro B. #startup #ShibuyaStartupSupport #Shibuya #Tokyo #event #founders #startupecosystem #YCombinator

The banks that will lead on operational efficiency over the next five years are not the ones with the largest technology budgets. They are the ones that stopped waiting for the perfect enterprise platform and started automating the workflows that are already fully defined. Every major bank has dozens of them. KYC document review. AML alert triage. Loan underwriting document routing. Contract clause extraction. Investment document assessment. Regulatory reporting preparation. The logic for every one of these workflows already exists. It is written down in compliance policy documents, operations manuals, and audit frameworks. The steps are defined. The decision rules are clear. The output requirements are known. The only thing missing is a tool built to run those workflows automatically, on-premise, with a full audit trail, in a way that both your compliance team and your regulator can inspect. That is what Jinba is built for. Not AI that replaces your compliance team. Not a six-month consulting engagement. Not a cloud platform that fails the security review. Deterministic workflow automation for the processes that govern how a Tier 1 bank actually operates. Deployed in days. Full audit logging at every step. Workflow logic your operations lead can read and verify without filing a ticket to engineering. The workflows are ready. The logic is already defined. The only question is how long you continue running them manually. Visit jinba.io or reach out directly to start the conversation.

Investment document review at major banks is still done by hand. Analysts reading through prospectuses, term sheets, and due diligence packages one by one. Flagging issues. Logging findings. Moving to the next document. The painful part is that the review logic is already defined. Every bank has a checklist. Every checklist has a decision tree. Every decision tree has a deterministic output. There is no ambiguity about what to look for or what to do when you find it. So why is a senior analyst spending hours on work a defined process could handle automatically? Because the tools available were never built for this environment. General-purpose automation platforms cannot handle the compliance constraints. They cannot produce the audit trail a regulator expects line by line. They were not designed for on-premise deployment inside a Tier 1 bank's infrastructure. And they require engineering resources most ops teams cannot access on short timelines. So the workflow stays manual. Document volume grows. Review cycles stretch. Analysts are pulled away from decisions that actually require judgment. Jinba automates investment document assessment workflows end to end. The workflow logic is written in plain language your Head of Operations and compliance lead can both read and sign off on. Full audit trail at every step. On-premise deployment. Deployed in days, not quarters. If your document review cycle is still running on analyst hours, the process is ready to be automated. The tooling just has to be built for it. Visit jinba.io to learn more.

At most major banks, contract review is still done by an analyst reading a document and flagging clauses manually. Not because the logic is unclear. The review criteria exist. The clause types are defined. The escalation rules are written down somewhere in a compliance policy document. It is manual because nobody has built a tool that can handle the full workflow — document ingestion, clause identification, logic application, compliance check, audit output — without a six-month IT project to get it running. So the queue grows. Senior analysts spend their hours on document intake instead of the decisions that actually require their judgment. Review cycles stretch. Counterparty onboarding slows down. Deals take longer to close. The bottleneck is not analytical capacity. It is the absence of automation in a workflow that is almost entirely automatable. Jinba handles the full contract review workflow as it actually exists inside a bank. Document ingestion from existing systems. Clause identification against defined criteria. Logic applied at every step with a full audit trail. Exceptions flagged for human review with complete context. Output formatted for compliance sign-off. Workflow logic written in plain language that your legal, compliance, and operations leads can all read and verify themselves. No black box. No six-month implementation. Deployed in days on your own infrastructure. Your senior analysts should be making decisions. Not reading documents that a well-defined workflow could have already screened. Visit jinba.io or reach out directly.