Kita: Turn documents into signals for lenders

Turn documents into signals for lenders

In emerging markets like the Philippines, open finance is still nascent. Most of the population is traditionally unbanked, banking APIs don’t exist, and a borrower’s financial history lives in documents: e-wallet records, bank statements, utility bills, and more. Because this data is unstructured, credit and risk teams are forced into manual review. This slows decisioning, increases costs, and caps lending volume. Legacy OCR solutions break on noisy, real-world files and still require human verification. Kita is the first document intelligence platform built specifically for lending. We are hyperlocalized around the signals that actually drive lending outcomes in emerging and undertapped domestic markets like the Philippines, Indonesia, Mexico, and beyond. Using a layered system led by vision-language models and computer vision, we outperform traditional OCR by transforming messy borrower documents into fraud-checked, decision-ready signals lenders can use directly in underwriting. Under the hood, Kita is a learning system. We link document-level signals to repayment outcomes, allowing our models to continuously improve fraud detection and risk assessment over time. This creates a compounding advantage for lenders as their distinct underwriting decisions feed back into the system. We’re Carmel and Rhea. We met before Stanford and have been building together ever since. Carmel is from Manila, is a repeat founder, and spent three years in product at Apple. Rhea has a research background in computer vision and received the highest honor in Stanford Computer Science. Together, we combine deep local context with strong technical execution to build the infrastructure that expands access to credit in emerging markets.

Active Founders

Carmel Limcaoco

Founder

Carmel is the co-founder and CEO of Kita. She is from Manila and studied Symbolic Systems & Music at Stanford. Prior to Kita, she was completing her Master’s in Computer Science. She spent three summers in audio and music product at Apple, where she was one of few interns to ever ship a feature in iOS. She was awarded by the United Nations at 16, launched the first Product Fellowship in the Philippines, and co-founded DAHA in 2022, where she took Stanford’s first marketplace from 0 to 1.

Carmel Limcaoco

Founder

Carmel is the co-founder and CEO of Kita. She is from Manila and studied Symbolic Systems & Music at Stanford. Prior to Kita, she was completing her Master’s in Computer Science. She spent three summers in audio and music product at Apple, where she was one of few interns to ever ship a feature in iOS. She was awarded by the United Nations at 16, launched the first Product Fellowship in the Philippines, and co-founded DAHA in 2022, where she took Stanford’s first marketplace from 0 to 1.

Rhea Malhotra

Founder

Rhea is the co-founder and CTO of Kita. Prior to Kita, Rhea completed her bachelor’s and master’s degrees in Computer Science at Stanford, and was an incoming PhD in Computer Vision and Robotics. She received the Firestone Medal, the sole highest honor of the Stanford CS Department, for her research in 2025. Rhea joined her first research lab at age 13, spent her gap year working at Pfizer on the COVID-19 vaccine, and has authored 7 papers recognized and awarded at international conferences.

Rhea Malhotra

Founder

Rhea is the co-founder and CTO of Kita. Prior to Kita, Rhea completed her bachelor’s and master’s degrees in Computer Science at Stanford, and was an incoming PhD in Computer Vision and Robotics. She received the Firestone Medal, the sole highest honor of the Stanford CS Department, for her research in 2025. Rhea joined her first research lab at age 13, spent her gap year working at Pfizer on the COVID-19 vaccine, and has authored 7 papers recognized and awarded at international conferences.

Company Launches

Kita: Turn financial documents into risk signals for lenders

See original launch post

Hi YC! We’re Carmel Limcaoco and Rhea Malhotra, the co-founders of Kita. We met over our gap year during COVID while we were undergrads at Stanford, and have been best friends ever since. Carmel is from Manila, is a past founder, and worked in computational music and audio at Apple. Rhea has a background in computer vision & robotics and was awarded the highest honor in Stanford Computer Science. We were both completing our Master’s in CS at Stanford when we chose to build Kita full time.

TLDR: We turn messy borrower documents into the data layer that powers lending in emerging and undertapped domestic markets.

Kita (YC W26) Launch Video

The problem

In emerging markets like the Philippines, where there are no banking APIs, a borrower’s financial history is locked in noisy documents: bank statements, payslips, utility bills, and more. For lenders, accessing and validating a borrower’s data is nearly impossible.

Today, lenders either review these documents manually or rely on legacy OCR. Manual review is slow and expensive, while OCR breaks on real-world inputs and still requires human verification. This slows loan decisioning, raises costs, and caps lending volume.

What is Kita?

Kita uses vision-language models to outperform traditional OCR, turning noisy borrower documents into fraud-checked, decision-ready signals for underwriting models. We are built specifically for emerging and undertapped markets, hyperlocalized around the signals that actually drive underwriting outcomes in places like the Philippines, Indonesia, Mexico, and beyond.

Over time, Kita links historical repayment outcomes to document-level signals, becoming a custom learning engine trained on a lender’s own data and decisions.

We’re starting in Southeast Asia with plans to expand to other emerging markets. We’re also exploring underserved segments in more developed markets, like parts of the U.S. where borrowers still rely on documents.

Our ask

Intros: We’d love introductions to lenders or fintechs operating in lending or credit, in emerging markets or in the U.S.
Founder wisdom: If you’ve built or scaled lending, underwriting, or risk infrastructure, we’d love to learn from you and compare notes.
Operators in financial services: If you work with document-heavy workflows, we’d love to understand the problems with turning documents into usable data.

uploaded image