Skip to content
View Vishwanath-06's full-sized avatar

Block or report Vishwanath-06

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Vishwanath-06/README.md

NETRUNNER_PROFILE

root@netrunner:~# define netrunner

netrunner (n.)  someone who jacks into the Net, bends it, and
walks out with what they came for.


in my case:     I jack into models and data. I train them, strip them
down, break them on purpose, and measure exactly when
they can still be trusted.


root@netrunner:~# whoami
vishwanath  //  handle: Vishwanath-06  //  class: NETRUNNER  //  mode: SOLO
CYBERDECKi7 U deck, plus a Kaggle T4 (16 GB VRAM) as my Sandevistan: short bursts of borrowed speed
CHROMEUndergrad, CSE (AI/ML). Neural nets, spiking nets, conformal guarantees, agents
CONTRACT LENGTH4 to 8 months per op. Scope it, ship it, write it up
DESTINATIONThe Moon. In my case, IEEE conference proceedings

Sandevistan status HUD: Kaggle T4 16GB, M2 deck linked


CYBERWARE_LOADOUT // RIPPERDOC INVENTORY

[ NEURAL_CORES ] [ CLOUD_UPLINKS ] [ CYBER_OPTICS ]
PyTorch
Python
snnTorch
Jupyter 		AWS
Kaggle 		TikZ
Data Viz

QUICKHACKS // INSTALLED DAEMONS


ICE_BREAKER // CURRENT_OPS

[ DAEMON ] [ EFFECT ] [ SOURCE ]
PULSE.exeReads heart rhythms and flags arrhythmiaPulsePod
BRIDGE.exeTranslates whole Indic-language documentsBhashaSethu
PRUNE.exeStrips weights until the model starts to hurtSparse-Neural-Nets
GHOST.exeRetrieval and generation with the uplink cutOffline_RAG
CALIBRATE.exeSets alarm thresholds you can trust, with conformal guaranteesUnder R&D 

root@netrunner:~# cat current_ops.txt

OP_01  BUGGED_CHROME        LLM agent swarm on local sys
OP_02  NIGHT_CITY_GRID      Dynamic graph transformers
OP_03  CYBERPSYCHOSIS_WATCH Conformal thresholds for visual anomaly
detection under drift (MVTec-AD)
OP_04  CHROME_CRASH_PREDICT Spiking nets for seizure prediction (MITBIH+PTB_XL)


root@netrunner:~# echo $TARGET
THE_MOON  // IEEE conference proceedings


root@netrunner:~# _

Open a file to read the briefing:

cat ops/OP_01_BUGGED_CHROME.log

Agents live and die by their tools. APIs time out, calls return garbage, schemas change mid-run. This op is about what an LLM agent swarm runnin g of U stage gig in parallel with Sandevistan.

cat ops/OP_02_NIGHT_CITY_GRID.log

Real networks rewire. Nodes join, edges decay, the topology you trained on is gone by tomorrow. Dynamic graph transformers are how I'm studying models that keep up with a grid that never holds still.

cat ops/OP_03_CYBERPSYCHOSIS_WATCH.log

Cyberpsychosis is a system drifting past its threshold without noticing. The research version: a visual anomaly detector on MVTec-AD with a frozen WideResNet-50 backbone, an alarm threshold set with conformal methods, and no labels to check it against. Can that threshold be trusted when the data drifts and nobody is annotating?

cat ops/OP_04_CHROME_CRASH_PREDICT.log

A spiking-network framework that predicts seizures on MIT-BIH & PTB-XL, like a ripperdoc's early warning before the chrome crashes the brain. Energy claims stay honest: I report SynOps and spike sparsity as simulation proxies and make no hardware-performance claims.


ACTIVE_BOUNTIES // PINNED GIGS

PulsePod: arrhythmia detection with deep and spiking neural networks BhashaSethu: document-level translation for low-resource Indic languages
SCC-ALAR: skin lesion classification with hybrid CNN-Transformers Sparse-Neural-Nets: pruning study, dense vs sparse
Offline_RAG: retrieval-augmented generation that runs offline FusionPhishGuard: A MultiBranch Deep Learning framework for Phishing Detection

OVERRIDE caution tape


STREET_CRED // NCPD BOUNTY BOARD

Vishwanath-06 GitHub stats Most used languages
Streak stats
Recent commit activity

OPEN_CHANNEL // JACK IN

Got a research idea, a collab, or a question about one of the ops? Pick a channel. Each button opens a pre-filled issue on this repo.

Collab Query Bug report


OVERRIDE caution tape

SYS.SHUTDOWN // "Built different."

Pinned Loading

  1. PulsePod PulsePod Public

    PulsePod is an AI framework for robust ECG arrhythmia detection. It unifies MIT-BIH and PTB-XL datasets to prevent dataset shift. Featuring high-accuracy 1D-CNNs and energy-efficient Spiking Neural…

    Jupyter Notebook 2

  2. FusionPhishGuard FusionPhishGuard Public

    Phishing detection, mobile and web security, attention mechanism, multi-branch deep learning, word embeddings, transformer, large language models (LLMs), BiLSTM, fusion framework.

    Jupyter Notebook

  3. BhashaSethu BhashaSethu Public

    BhashaSetu (Sanskrit for "Language Bridge") is an AI-powered document-level machine translation workspace designed specifically for low-resource Indic languages. Built on IndicTrans2, it provides s…

    Python 1

  4. Sparse-Neural-Nets Sparse-Neural-Nets Public

    PyTorch experiments on neural network sparsity and pruning. Compares dense and sparse models on Breast Cancer (FCNN) and MNIST (CNN) using L1, structured, and random pruning across multiple sparsit…

    Jupyter Notebook 2

  5. SCC-ALAR SCC-ALAR Public

    SCC-ALAR: An experimental study on skin lesion classification using hybrid CNN-Transformer architectures. This project explores the impact of architectural complexity versus inductive bias on the i…

    Jupyter Notebook 1

  6. Offline_RAG Offline_RAG Public

    A lightweight, fully offline Retrieval-Augmented Generation (RAG) system using LangChain, ChromaDB, and local embeddings. It enables secure document processing and question answering entirely on y…

    Python 1