I build things that ship — then I sell them.
CS student at Chennai running three tracks at once: systems engineer · ML researcher · digital entrepreneur. Python-first, comfortable from GPU shaders to Rust systems to production ML.
I've built and sold 7 digital products — all launched organically through Reddit & Discord. No VC, no paid ads.
The edge: I write the code, design the architecture, train the model, run the campaign, and close the sale. Most people pick one — I compound all of them.
Primary: Python · FastAPI · PyTorch · React · Three.js · AWS
Systems: C++17 · Rust · WebAssembly · GLSL
Open-Source: SYNTHRON · VORTEXRAG · puredata.py · datamend.py
Python Dev My primary stack. ML pipelines · FastAPI · automation · data tools · CLI libraries. Where I'm fastest. |
ML Architect Novel transformer architectures. Graph topology · temporal decay · adaptive compute. Not fine-tuning — designing. |
WebGL Wizard Three.js · GLSL shaders · WebAssembly · 600K GPU particles in-browser. Pushing what browsers can do. |
Growth Hacker Google Ads · Meta · SEO · Reddit launches · Discord communities. Zero ad spend, organic conversions. |
Entrepreneur 7 digital products built and sold. eLearning · dev tools · automation. Different platforms, different prices. |
Cloud Optimizer AWS + GCP cost optimization. Avg 30-50% infra reduction. Grafana monitoring. Startups love this. |
AI Framework Builder SYNTHRON multi-agent framework · VORTEXRAG 7-layer RAG · LLM eval · agentic orchestration at scale. |
Mobile Engineer Flutter · Firebase · Riverpod. Built SparshCare — 50-file ICU patient communication app, 4 languages. |
Open Source Creator 40+ public repos. puredata.py · datamend.py · SYNTHRON · VORTEXRAG. Real libraries solving real pain. |
Data Viz Expert CesiumJS · Three.js · D3.js · Grafana. FLY-VIEW globe · Graphite metrics · Tradeorbit trade flows. |
LLM Benchmarker Production LLM eval framework. GPT-4 · Claude · Gemini · Mistral. Accuracy, latency, cost, hallucination. |
"The best engineers I know don't just write code — they ship things people pay for."
7 digital products. Built solo. Sold through community. Zero VC. Zero paid ads.
Flagship: a full eLearning platform — sold for 68K. Course infra, payments, progress tracking, instructor dashboards. Marketed and sold entirely through Reddit & Discord. Zero ad spend.
Product Portfolio
| # | Category | Type | Price Range |
|---|---|---|---|
| 1 | eLearning Platform | Full SaaS | 68K |
| 2 | Developer Automation Suite | Tool | 12K – 18K |
| 3 | AI Analytics Dashboard | SaaS | 20K – 35K |
| 4 | Content Creator Toolkit | App | 5K – 8K |
| 5 | Resume & Portfolio Builder | SaaS | 8K – 15K |
| 6 | Community Management Bot | Tool | 6K – 10K |
| 7 | Data Pipeline Automation | B2B Tool | 25K – 40K |
Launch platforms: Reddit (primary) · Discord servers · Slack workspaces · Developer forums
|
Startups & Scale-ups Senior-level engineering without senior-level headcount — full-stack features, ML pipelines, cloud infra, WebGL/3D data products. |
HR & Talent Teams Internal tools — ATS integrations, automated screening, candidate analytics, attendance intelligence. |
|
Cloud Cost Optimization — AWS + GCP Audit, right-size, monitor via Grafana + CloudWatch. Average 30–50% infra cost reduction, zero performance loss. |
Digital Marketing for Tech The bridge between founders who can't market and marketers who can't speak to engineers. SEO + community-led growth that converts technical audiences. |
Star the Repo · Docs · npm · Get Pro |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Open-source. Live. The RAG architecture the field is missing.
VORTEXRAG is now public. A 7-layer production RAG framework that solves what standard pipelines can't: semantic drift and context poisoning — simultaneously. Built on the TemporalMesh Transformer backbone with custom retrieval that understands temporal semantic relationships, not just cosine similarity.
STANDARD RAG: Query -> Embed -> Top-K -> Stuff context -> Generate
Problem: No temporal awareness. Context poisoning. Drift.
VORTEXRAG: Query -> Temporal Embed -> Orthogonal Resonance Tuning
-> 7-Layer Extraction -> Confidence-Gated Context -> Generate
Result: The only RAG that kills semantic drift + context poisoning
| Layer | Function | Status |
|---|---|---|
| Vector Orthogonal Tuning — noise cancellation on embedding space | Core innovation | LIVE |
| Resonance Retrieval — semantic resonance, not cosine similarity | Core innovation | LIVE |
| Temporal Decay — time-aware context weighting | Core innovation | LIVE |
| Context Poison Detector — filters adversarial/stale context | Safety layer | LIVE |
| Confidence Router — skips generation if retrieval confidence low | Reliability layer | LIVE |
| Aerospike Vector Store integration | Storage layer | IN PROGRESS |
| FastAPI streaming interface | API layer | IN PROGRESS |
Two libraries. Both built from real pain points across 5+ projects. On GitHub now.
puredata.py — Automatic Data Cleaning & Drift Detection
Nulls sneak in. Distributions shift. Models fail silently on corrupted inputs. puredata is the library that catches it before it breaks your pipeline. Zero config, drop-in.
import puredata as pd
df = pd.clean(your_dataframe) # auto-cleans nulls, outliers, type mismatches
pd.monitor(df, interval=60) # silent drift detection in productiondatamend.py — Production Data Quality
The single library that catches, fixes, validates, monitors, and traces every data quality issue — automatically. So your ML pipeline never breaks from bad data again.
from datamend import Pipeline
pipeline = Pipeline(df)
pipeline.fix().validate().monitor() # one chain, complete data qualitypuredata.py on GitHub · datamend.py on GitHub · Announced on Reddit and Discord
Every product I've sold started with a Reddit post, a Discord message, or a Slack thread.
I don't launch behind paywalls or in isolation. I build in public, share progress, and let the community validate before I optimize. Free launches first, real feedback always, paid version after trust is built.
Why Reddit works: Technical communities have the highest-intent buyers on the internet. They evaluate rigorously, tell their networks when something is good, and prefer buying from builders they've seen in the community.
|
|
|
|
|
|
|
VIGNESHWAR · Engineer · Entrepreneur · ML Architect · CS @ Chennai
Open to collabs · contracts · co-founders · Respond within 24h
