Hello, World. I am

Jithendra Nara

>_ █

Architecting high-performance systems at the intersection of complex data, scalable software, and artificial intelligence. I bridge the gap between algorithmic research and production-grade engineering to solve intractable problems.

View Work Initiate Contact

Python

React

AI/ML

Spark

Beyond The Editor

My engineering philosophy is rooted in systems thinking. Whether I'm designing a data lakehouse architecture or orchestrating multi-agent AI ecosystems, I believe in building resilient, scalable foundations that empower product velocity.

I don't specialize in a single layer of the stack — I architect across it. From the ingestion layer and feature stores to the inference API and the UI that renders the result, I understand how each decision propagates through the system.

But life isn't just about the terminal. I'm a passionate creator and competitor outside of the codebase.

Photography Formula 1 Cricket & Soccer Testing AI Models Vibecoding

Years Engineering

GitHub Repos

Cloud Platforms

∞

Problems to Solve

Core Domain Matrix

Languages

Python

TypeScript

JavaScript

SQL

Swift

Kotlin

Shell

Data & Analytics

Spark

Pandas

Snowflake

dbt

Tableau

Airflow

Intelligence & AI

TensorFlow

PyTorch

LangChain

OpenAI

Gemini / ADK

NVIDIA NIM

Architecture & Backend

FastAPI

Node.js

Express

Flask

REST APIs

GraphQL

Infrastructure

Docker

K8s

AWS

GCP

DigitalOcean

Terraform

Frontend & Mobile

React

Next.js

Three.js

Tailwind

D3.js

Vite

Tech Ecosystem

Everything I've shipped production code with — distilled from 78 repositories and counting

AI / LLM Frameworks

Google ADK

LangChain CrewAI

LlamaIndex LlamaParse

NVIDIA NIM

Pinecone RAG Pipelines

OpenAI API Gemini API MiniMax

ML / Deep Learning

TensorFlow

PyTorch

Scikit-Learn TensorFlow.js

MLflow PoseNet KNN / Gradient Boosting

Data Engineering

Apache Spark

Airflow

dbt Delta Lake

Snowflake Medallion Architecture ETL / ELT Pipelines

Redshift AWS Glue

Cloud & DevOps

AWS (S3, Lambda, Glue)

GCP (Cloud Run)

DigitalOcean

Vercel

Docker

Kubernetes

Terraform

GitHub Actions

Nginx Gunicorn

Databases & Storage

PostgreSQL

SQLite

Oracle

Pinecone

S3 Vector Stores

Frontend & Visualization

React

Next.js

Three.js

D3.js WebGL

Tableau

Streamlit

Tailwind Chart.js GSAP

Monitoring & Tooling

Sentry

Git systemd

Telegram Bot API

Discord Bots Codex CLI OpenClaw

Architecture Showcase

Click any project for a full case study

Personal Assistant ADK

Production multi-agent AI system on Google ADK with 7 specialist agents, 6-stage callback pipeline, memory archival, and identity injection.

Google ADK

Python

Docker

Cloud Run

Source

CricBoxd

The Letterboxd for Cricket — a social diary to log matches, rate innings, write reviews, and discover the greatest games ever played.

JavaScript

Node.js Social Platform

Source

Analytics Copilot RAG

RAG-powered analytics assistant grounded in project docs, schemas, and KPI definitions with guardrailed SQL suggestions.

Python

LangChain RAG SQL Guardrails

Source

Autonomous Research Agent

Serverless multi-agent research assistant on GCP Cloud Run. Planner, Search, Synthesis, and Critic agents. $0.003/task.

Next.js

Python

CrewAI

GCP

Source

Casino Analytics

Behavioral analytics platform for casino player data — ETL pipelines, churn prediction ML model, real-time FastAPI endpoints, and Tableau dashboards.

Python

Snowflake

Tableau

FastAPI

Source

LlamaParse + MiniMax

PDF parsing pipeline using LlamaCloud for document extraction and MiniMax M2.7 for plain-English financial document summarization.

Python

LlamaIndex MiniMax LlamaParse

Source

RAG Document Detective

Semantic search over 50K+ documents. Pinecone vector store, OpenAI embeddings, cross-encoder reranking. <2s query response.

Next.js

Pinecone

OpenAI

LangChain

Source

Lakehouse Analytics

Medallion architecture (Bronze-Silver-Gold) on Delta Lake. Spark Structured Streaming, Airflow DAGs, dbt. Processes 10TB+ daily.

Spark

Airflow

dbt Delta Lake

Source

FrailtyTester

Digital health assessment platform with TensorFlow.js PoseNet for AI-powered physical movement analysis via webcam. 94% accuracy.

React.js

Node.js

TensorFlow.js

Source

Tollywood Chronicles

Immersive 3D data visualization of 26 years of Telugu cinema. Three.js + D3.js + WebGL. Interactive, filterable, 60fps.

Three.js

D3.js WebGL

Source

Open Source Footprint

Merged PRs, bug fixes, and feature contributions to projects I use daily

Projects Contributed To

Pull Requests

Merged

Issues Filed

OpenClaw

36.9k

Merged

Added MiniMax as a bundled web search provider, including native support through the Coding Plan search API with global and CN endpoint routing.

feat(tools) size: L extensions: minimax

PR #54648

Repo

Also filed four issues around Telegram streaming, CLI provider behavior, and proactive context compaction.

Hermes Agent

37k

Open

Fixed an async coroutine leak where unscheduled coroutines in thread-safe bridges were never properly closed, causing resource exhaustion.

fix(async) NousResearch

PR #2657

Repo

Google LangExtract

Google

Open

Added MiniMax model support through an OpenAI-compatible API, extending Google's document extraction library with another LLM provider path.

feat size: L

PR #416

Repo

Google Stitch SDK

Google Labs

Merged

Added README badges and exported `StitchProxyConfigSchema`, improving discoverability and day-to-day developer experience for the SDK.

feat docs

PR #110

Repo

MiniMax AI Skills

9.6k

Merged x2

Two merged PRs for plugin metadata sync and install docs, plus two open bug-fix PRs for the DOCX skill build pipeline.

chore docs fix

PR #26 PR #25

Repo

OpenClaude

CLI Agent

Open

Fixed model-selection logic so initial provider environment settings are respected during model picker initialization.

fix multi-provider

PR #306

Repo

MiniMax Coding Plan MCP

MiniMax AI

Open

Removed overly aggressive steering from tool descriptions so the MCP server stops over-constraining agent behavior with opinionated schema prompts.

fix MCP

PR #17

Repo

View Full GitHub Profile

Engineering Notes

Short field notes on agentic systems, retrieval, data platforms, and the parts of shipping that do not fit inside a project card

Apr 2026

What Open Source Taught Me About Agentic Systems

Lessons from fixing provider routing, coroutine leaks, tool descriptions, and memory surfaces across active AI agent repositories.

Read note

Apr 2026

RAG Systems Beyond the Demo

Why the hard part is not the vector database. It is evaluation, permissions, schemas, freshness, and knowing when not to answer.

Read note

Apr 2026

Static Portfolios Should Be Feedback Machines

A static site can still tell you what works: section depth, outbound clicks, contact starts, successful submissions, and dead zones.

Read note

View All Notes View Learning Timeline

Initiate Contact

Open to full-time roles, contracts, and research collaborations

[email protected]

github.com/JithendraNara

jithendranara.dev