Abhishek PatelPortfolio
Build.
000
Work
AI / RAGApr - Jun 2025Private client

PiSolved Chat

A solo, end-to-end RAG chatbot with a ChatGPT-like experience over domain knowledge.

Sole end-to-end developerVedteq (Oman)
Private client
Next.jsTypeScriptOpenAI Assistants APIPostgreSQLDrizzle ORMResend
Visuals under NDA

PiSolved Chat is a private client engagement. The live platform is the proof; implementation details are available to walk through under NDA.

01The problem

Users needed domain-specific answers with the fluency of ChatGPT, but grounded in a specific knowledge base - with persistent threads and sensible usage limits.

02The solution

An end-to-end RAG chatbot I built entirely solo: real-time streaming chat, thread-based persistent sessions, daily quotas, and an email workflow for quota upgrades.

03Architecture
Chat

Real-time streaming interface, thread-based persistent sessions

AI

Context-aware responses via the OpenAI Assistants API

Data

Full schema in PostgreSQL with Drizzle ORM

Ops

Daily quota management + Resend email workflow for upgrade requests

04Key features

Streaming chat

Real-time token streaming with auto-scroll and loading states.

Persistent threads

Thread-based sessions that survive reloads.

Quota system

Daily quota management with a Resend-powered upgrade flow.

Grounded answers

Domain-specific responses via the Assistants API.

05Results
MetricValue
solo build0-7 wk
end to end, solo0%
experience at the timeUnder 1yr
grade deliveryProd
06My role

I built this entirely solo in 6-7 weeks: the streaming chat UX, thread persistence, the OpenAI Assistants integration, the full PostgreSQL/Drizzle schema, the daily quota system, and the Resend email workflow. This was the project that confirmed my transition from frontend to genuinely full-stack.

07What I learned
  • RAG quality is mostly retrieval and grounding discipline, not model choice.
  • Quotas and upgrade flows are part of the product, not an afterthought.
08FAQ

Want the deeper story?

Ask my AI about PiSolved Chat for specifics, or reach out directly.

Get in touch
Next project
PiSolved Platform