Open to opportunities

Sridharan T

Full StackEngineer

I ship AI agents, ONNX model servers, and real-time systems — LangGraph orchestration, FastAPI endpoints, VS Code extensions, and everything in between. Full Stack Engineer at Aspire Systems (50K+ users).

terminal
~/social-agent $

50K+

Users served

on live platforms

40%

API latency reduced

at Aspire Systems

15+

Public repos

AI · tooling · web

PythonTypeScriptLangGraphNext.jsFastAPIONNXDocker

The person behind
the code

Full Stack Engineer at Aspire Systems building AI-powered products used by 50K+ users. I specialize in Python, TypeScript, and LLM integrations — from LangGraph agents and ONNX model servers to real-time chat systems and VS Code extensions. I ship production code daily and explore the edges of what AI-powered tooling can do.

012+ years shipping production systems at Aspire Systems (50K+ users)
02Built LangGraph agents, ONNX embedding servers, and VS Code extensions
03Fluent across the full stack: Python · TypeScript · PHP · Node.js · React
04Integrates OpenAI, Anthropic, Google Genkit, and ONNX into real products
05Deployed on AWS EC2, Firebase, Vercel, Docker, and Railway
06Open-source contributor with 15+ public repos across AI, tooling, and web

Bachelor of Technology in Information Technology

Bannari Amman Institute of Technology

July 2019 – April 2023

LocationSalem, Tamil Nadu, India

2+

Years at Aspire

50K+

Users impacted

15+

Public repos

40%

Perf. improvement

Career journey

From structured training to shipping production features for 50K+ users — growing role and responsibility at each phase.

Years at Aspire2+
Users impacted50K+
Storage cost saved30%
API latency reduced40%
Current

Software Engineer

Aspire Systems

Apr 2024 – Present
  • Led core development of an AI-enhanced email marketing platform serving 50K+ active users
  • Architected RESTful APIs for campaign management and real-time analytics using Node.js and MongoDB

Trainee Engineer

Aspire Systems

Jan 2024 – Mar 2024
  • Contributed backend service development in Node.js and Laravel for production platform features

IMPACT Trainee

Aspire Systems

Jan 2023 – Mar 2023
  • Completed structured full-stack training covering JavaScript, Node.js, MongoDB, and REST APIs

Technical expertise

Languages

Python95%
TypeScript90%
JavaScript92%
PHP80%

Frontend

React / Next.js90%
Vue.js82%
Angular75%
Tailwind CSS95%
Framer Motion85%

Backend & AI

Node.js / Express92%
Laravel / PHP80%
FastAPI85%
LangGraph / LangChain88%
OpenAI / Anthropic90%
ONNX Runtime80%

Database & Cloud

MongoDB90%
PostgreSQL / Prisma82%
Firebase85%
AWS (S3, EC2)82%
Docker85%
Socket.IO / Pusher88%

Crafted to solve real problems

01

Social Media Agent

A LangGraph-powered autonomous agent that simultaneously posts to LinkedIn, Twitter/X, Facebook, and Instagram — adapting tone, length, and hashtag strategy per platform using GPT-4 and Claude.

PythonLangGraphLangChainOpenAIAnthropic
  • Orchestrated parallel platform posting via LangGraph StateGraph with memory checkpointing
  • AI rewrites content per platform: 280-char Twitter punchy vs long-form LinkedIn professional
  • Supports media uploads, error retry logic, and cross-platform status tracking
  • Deployed via Docker, Fly.io, and Railway with environment-aware configuration
02

FocusFlow — AI Task Manager

An AI-powered task management PWA built for ADHD users. Google Genkit intelligently schedules and prioritizes tasks based on cognitive load and current time, with real-time push notifications on iOS.

Next.js 15FirebaseGoogle GenkitPusher
  • Google Genkit AI understands current datetime to generate context-aware task schedules
  • Real-time cross-device notifications via Pusher + Web Push API (Safari/iOS support)
03

AI Code Assistant — VS Code Extension

A VS Code extension providing an intelligent chat interface with Ghost Chat — a cursor-positioned floating AI panel that shows diffs inline, like Cursor IDE, before any change is applied.

TypeScriptVS Code APIWebViewMonaco Editor
  • Ghost Chat: Ctrl+K triggers cursor-positioned chat overlay with diff-style previews
  • Drag-and-drop file attachment from explorer, tabs, and system files into chat context
04

ONNX Embedding Server

A portable, offline-capable embedding server using FastAPI and ONNX Runtime. Packaged as single-binary executables for Linux, Windows, and macOS with the model bundled — no internet, no Python install required.

PythonFastAPIONNX RuntimeDocker
  • Serves sentence-transformers/all-MiniLM-L6-v2 via REST API — fully offline
  • Model + tokenizer bundled into a 22MB standalone binary using Nuitka

Open Source Activity