Darsh Patel DARSH // ONLINE
> BOOTING SYSTEM...
00%
Darsh Patel
AI & Data Science · IIT Jodhpur

Darsh Patel

Exploring how intelligent systems are built.

I build intelligent software: agents, developer tools, and the systems underneath. Recently, I shipped Oracle’s database provider for Mastra.

github.com/darshpatel1052
Bitonic network sort
01 About

About

Current Interests

LLMs Agentic Architectures Distributed Systems System Design Databases Open Source Algorithms

I'm a final-year AI & Data Science student at IIT Jodhpur with a strong interest in the engineering behind intelligent software. What began as curiosity about machine learning has gradually evolved into a broader fascination with the systems, abstractions, and design principles that make modern AI possible.

I enjoy connecting ideas across LLMs, distributed systems, databases, and software architecture. Whether I'm building projects, contributing to open source, or exploring a new topic, I'm motivated by understanding how things work before thinking about how they can be improved.

02

Experience

MAY — JUL 2026 · recent

Software Engineering Intern · Oracle

Open-source database infrastructure

Shipped @mastra/oracledb, a first-party Oracle Database provider for Mastra. Persistent memory, workflows, observability, agents, and MCP clients — with AI Vector Search underneath.

3,000+ monthly npm downloads640+ integration testsExact, HNSW & IVF retrieval
Explore the contribution ↗
FEB — APR 2025

Software Developer Intern · Reetape Technologies

Real-time voice systems

Built a real-time voice customer-service agent with local Ollama inference, streaming responses, and speech synthesis. Connected the browser to the conversation with Next.js and WebRTC.

< 200 ms client-side latency~5× faster backend responses
2024–25

Research Intern · IIT Hyderabad

Computer vision · generative models

Conducted research on controllable video weather translation using deep generative models, focusing on transforming scenes across weather conditions while preserving temporal consistency, motion, and visual realism.

Computer VisionDiffusion ModelsGANsPyTorch
Research · Ongoing

Improving Small Vision-Language Models through Data-Centric Learning

Exploring how small vision-language models can recover large-model capabilities by learning from high-quality, filtered supervision instead of larger datasets. Current work focuses on trust-aware pseudo-label filtering, hidden-state based verification, and improving data quality for efficient model adaptation.

Vision-Language ModelsData-Centric AIPseudo-Labeling
Beyond the code
Top 100Amazon HackOn
Amazon ML Summer SchoolSelected from 60,000+ applicants
Teaching AssistantIntro to Computer Science · IIT Jodhpur · 2025
02 A few things I've enjoyed building.
03

Selected Work

Browser extension · RAG03

Askify↗

Contextual AI assistance inside your browser. Index a page, ask a question, and get an answer grounded in its content — without leaving the tab.

Chrome ExtensionLangChainRAG
Page context · Conversation historySource ↗
All projects on GitHub ↗
04

Stack

/ 01 Languages

CC++GoJavaPythonTypeScriptJavaScriptSQL

/ 02 AI & Agent Systems

AI AgentsLLMsOllamaMastraLangGraphGoogle ADKLangChainHugging Face

/ 03 Backend & Web

Next.jsReactNode.jsFastifyFastAPIgRPCREST APIsWebRTCChrome Extensions

/ 04 Databases

Oracle DatabasePostgreSQLpgvectorRedisKafkaBullMQMongoDBChromaDBSupabaseFirebase

/ 05 Infrastructure & Tools

GitGitHubDockerTerraformGitHub ActionsOpenTelemetryEnvoyAWSGoogle CloudPostman
05

Writing

Things I want to write down

Things I’m figuring out, projects I can’t stop thinking about, and the occasional tangent.

All posts ↗
06 "The tools change.
The fundamentals don't."

Let's talk.