Cloud-native full-stack products on AWS — now AI-enabled.

type help for commands — or just ask anything.

$
ready·0 cmds
⚡ edge FRARAG · pgvector
Full-Stack Developer • AI Apps, RAG & Cloud

Building AI-enabled full-stack products that stay useful in production.

I'm Kubilay Uysal, a full-stack developer focused on TypeScript, Next.js, PostgreSQL, and AWS. Lately I've been leaning harder into AI applications: RAG systems, document-based chat, pgvector-backed retrieval, and product experiences that connect useful model output with reliable backend execution.

Full-Stack AI AppsTypeScriptPostgreSQL + pgvectorRAG PipelinesDocument ChatAWS + SSTAWS Strands

Delivery profile

AI product UI + cloud delivery

Open to opportunities

Primary focus

AI-enabled full-stack products

Recent build

AskMyDocs

Data layer

PostgreSQL, pgvector, retrieval

AI applications

Productizing RAG, document chat, and model-assisted workflows into usable interfaces.

Retrieval layer

PostgreSQL, pgvector, ingestion pipelines, and grounded answers over raw generation.

Cloud delivery

AWS, SST, serverless APIs, and the operational discipline needed to ship real products.

Learn how I work

Profile

Kubilay Uysal

TR / Remote
KU

Focus

Full-stack engineering with strong TypeScript foundations and growing emphasis on AI applications, RAG, and document intelligence.

Recent build

AskMyDocs

Document-based chat with retrieval, pgvector search, and grounded AI responses over uploaded knowledge.

About

I'm a full-stack developer moving deeper into AI-native product work.

My work sits between product experience and system design. I still care about strong interfaces and clean backend structure, but the direction I'm excited about now is AI products: document intelligence, retrieval-backed chat, pgvector-powered search, and full-stack systems where the model layer is useful because the rest of the architecture is reliable.

AI product thinking

I like AI features that solve a real knowledge or workflow problem, not novelty demos that collapse outside a controlled prompt.

Retrieval architecture

RAG, document ingestion, chunking, embeddings, and PostgreSQL + pgvector feel like practical foundations for grounded answers.

Full-stack execution

I want the same level of care across interface, API, storage, eval loops, and deployment instead of treating AI as a disconnected layer.

Current direction

I'm positioning myself for international full-stack roles where AI product development, retrieval-backed systems, and cloud execution all matter at the same time. AskMyDocs pushed that focus forward for me in a practical way.

Experience

The work I want to be hired for now.

Rather than listing placeholder job history, this section reflects the stack and problem space I'm actively leaning into: AI application engineering, retrieval-backed systems, and full-stack delivery that is useful in production rather than only impressive in a demo.

01

AI application development

Turning model capabilities into usable full-stack products with clear interfaces, grounded responses, and workflows people can actually trust.

RAG appsPrompt flowsAI UX
02

Retrieval and knowledge systems

I'm especially interested in document ingestion, embeddings, PostgreSQL + pgvector, and the data design behind document-grounded chat.

pgvectorDocument chatGrounded answers
03

Cloud-ready full-stack delivery

Next.js, TypeScript, AWS, SST, and newer AI-oriented workflows like Strands matter to me when they help ship something stable and maintainable.

Next.jsAWS / SSTStrands workflows
Projects

Recent work and the systems I want to build more of.

AskMyDocs is the clearest signal of where I'm heading: AI products with real retrieval, useful interfaces, and backend structure that supports the model layer instead of fighting it.

Start a conversation

Recent project

AskMyDocs

AI focus

A document-based chat application where uploaded knowledge becomes searchable and answerable through retrieval instead of generic model recall.

Next.jsTypeScriptPostgreSQLpgvectorRAG

Document chat

Vector retrieval

Grounded answers

Active focus

AI knowledge interfaces

AI focus

Full-stack products that turn documentation, PDFs, and internal knowledge into conversational flows people can actually use.

Document ingestionChunkingEmbeddingsSearch UX

Lower search friction

Contextual answers

Practical AI UX

Platform layer

Cloud-native AI delivery

AI focus

Deploying AI-enabled applications with the same rigor as any other product: auth, API boundaries, storage, observability, and cost awareness.

AWSSST IonServerless APIsStrands experiments

Reliable release flow

Operational clarity

Scalable foundations

Full-stack capability

Developer tooling and product surfaces

AI focus

Internal tools, admin flows, and polished product surfaces that support the AI or backend layer instead of feeling disconnected from it.

Next.jsshadcn/uiRole-aware UIPostgreSQL

High-signal interfaces

Faster operations

Clean systems thinking

Contact

Looking for a full-stack developer with growing AI product depth?

I'm open to full-stack opportunities where AI applications, retrieval-backed systems, cloud architecture, and thoughtful frontend execution all matter.

Or write to me right here