Full-stack & AI systems engineer

KrishnaJadhav

I build web applications, backend services and automation systems — and the retrieval pipelines that make them useful.

LocationMumbai, India
Coordinates19.0760° N · 72.8777° E
FocusAI systems · Full-stack
Repositories13 public projects
Scroll2026
01/Capabilities

What I work on

Four areas I keep returning to, and the work each one covers.

01

Full-Stack Engineering

Next.js and React front-ends over Node, Express, Flask and Django services, with PostgreSQL and MongoDB behind them.

02

AI & Automation

Retrieval-augmented systems, vector search and multi-API agent pipelines orchestrated with n8n and hosted LLM providers.

03

Data & Machine Learning

Embedding pipelines, classical NLP models and exploratory analysis in Python, from notebook to deployed service.

04

Open Source

Public repositories for every project below — virtualised React components, CLI tooling and automation workflows.

02/Selected projects

Systems, pipelines and interfaces

Each entry links to its live deployment where one exists, its source, and a written case study covering architecture, problem and approach.

  • 01
    Systems Architect & Automation Engineer
    Active Production

    Fully automated WhatsApp bot for managing availability. Features sleep mode, busy status, and smart auto-replies using n8n & PostgreSQL.

    n8n WorkflowsMeta Cloud APIPostgreSQLNode.jsExpressDocker
  • 02
    Full-Stack Engineer & AI Developer
    Production Live

    Portfolio with RAG-powered semantic search. Visitors can query experience using natural language. Built with Next.js and Pinecone.

    Next.jsTypeScriptPineconeGoogle AI EmbeddingsTailwindCSSFramer MotionMDX
  • 03
    Full-Stack Developer
    Production Deployed

    Full-stack academic management platform for Rajarshi Shahu Mahavidyalaya. Features student enrolment, admin dashboards, and GridFS storage.

    FlaskPythonMongoDBGridFSREST APIRBAC Auth
  • 04
    Full-Stack AI Developer
    Shipped

    AI medical research platform retrieving papers across PubMed, OpenAlex, and ClinicalTrials.gov, synthesizing evidence-graded answers.

    ReactNode.jsExpressMongoDB AtlasGroq (Llama 3.3 70B)ONNX EmbeddingsRRF Re-ranking
  • 05
    Automation Engineer
    Active Automation

    Fully automated video generation pipeline. Orchestrates n8n, Gemini, Edge TTS, and MoviePy to turn a topic into a YouTube-ready MP4.

    n8n WorkflowsPythonGemini AIEdge TTSMoviePyFFmpeg
  • 06
    Full-Stack AI Developer
    Shipped

    AI-assisted coaching and performance analytics platform for tracking user progress and personalized skill development.

    ReactNode.jsExpressTailwindCSSChart.jsREST API
Showing 06 of 13
03/Profile

How I work

I’m a full-stack engineer and AI developer. Most of what I build sits at the seam between an interface and a system that has to keep running without me watching it.

That means retrieval pipelines that stay correct, workflow automations that recover from failure, and front-ends that hold up under real data. I care about clean boundaries and about code the next person can read.

01

Architecture over hype

Clear boundaries, strict types and production-grade error handling instead of chasing transient stacks.

02

Deterministic AI

LLMs, embeddings and multi-API pipelines wrapped in structured outputs, validation and retries.

03

Observable & maintainable

Self-documenting code, structured logging and handoff notes so the next engineer is never guessing.

04/Core technologies

The toolset

The stack I reach for first, grouped by where it sits in a system.

AI & Automation
LangChainLangGraphn8nPineconeGoogle AI EmbeddingsONNX RuntimeGroqGemini
Frontend
TypeScriptReactNext.jsVue.jsTailwindCSSFramer MotionMDX
Backend & Data
Node.jsExpressPythonFastAPIFlaskDjangoPostgreSQLMongoDB
Infrastructure
DockerGitLinuxVercelFFmpegCI/CD
05/Writing & work
06/Contact

Start a conversation

Send a note about a role, a build, or something you'd like to take apart together.

Direct
jadhavkrishna475@gmail.com
LocationMumbai, India
AvailabilityRemote worldwide
ResponseUsually within a day
Networks