Saran Karthick (saran612) | Portfolio & Projects

Saran
Karthick

b.Tech Student • Machine Learning Engineer

I engineer the underlying logic of the new internet. Focused on Machine Learning, Intelligent Automations, and crafting seamless human-computer experiences.

The Projects

def __init__(self, in_features, out_features, bias=True):
    super(Linear, self).__init__()
    self.in_features = in_features
    self.out_features = out_features
    self.weight = Parameter(torch.Tensor(out_features, in_features))
    
    if bias:
        self.bias = Parameter(torch.Tensor(out_features))
    else:
        self.register_parameter('bias', None)
        
def forward(self, input):
    return F.linear(input, self.weight, self.bias)

Project 01

Metro-Mind AI

A decision-support dashboard and real-time train scheduling for the Kochi Metro.

Real-time Scheduling Efficiency:99.98%

ML Predictive Accuracy:96.4%

System Load Optimization:-420ms

import cv2
import numpy as np

def process_frame(frame, threshold=0.8):
    tensor = transform(frame).unsqueeze(0)
    with torch.no_grad():
        outputs = model(tensor)
    return filter_boxes(outputs, threshold)

Project 02

Memora

A Retrieval-Augmented Generation (RAG) for personal knowledge management capabilities.

Object Detection Latency:12ms

False Positive Reduction:45.2%

Concurrent Streams:1,200+

function verifyTransaction(signature: string, payload: any): boolean {
  const hash = crypto.createHash('sha256').update(JSON.stringify(payload)).digest('hex');
  return crypto.verify('sha256', Buffer.from(payload), pubKey, Buffer.from(signature, 'hex'));
}

Project 03

AgriQCert

A end-to-end platform for a digital product passport system for agricultural imports and exports.

Cryptographic Verification:<1ms

Throughput (TPS):85,000

Consensus Finality:400ms

func (c *Cache) Get(key string) (interface{}, bool) {
    c.mu.RLock()
    defer c.mu.RUnlock()
    item, found := c.items[key]
    if !found || item.Expiration > 0 && time.Now().UnixNano() > item.Expiration {
        return nil, false
    }
    return item.Object, true
}

Project 04

Dark-Cloud

A social platform for connecting like-minds around the world.

Cache Hit Ratio:99.1%

Memory Overhead:1.2MB

Response Time (p99):0.8ms

SELECT
  user_id,
  COUNT(*) as events,
  APPROX_QUANTILES(duration_ms, 100)[OFFSET(95)] as p95_duration
FROM event_logs
WHERE timestamp >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 1 HOUR)
GROUP BY user_id;

Project 05

Arise

An interactive productivity tracker web app.

Events Processed/Sec:4.5M

Query Acceleration:14x

Storage Compression:82%

app.use(async (ctx, next) => {
  const rateLimit = await checkRateLimit(ctx.ip);
  if (!rateLimit.allowed) {
    ctx.status = 429;
    ctx.body = 'Too Many Requests';
    return;
  }
  await next();
});

Project 06

Drift

A hybrid learning platform combining AI and interactive content.

Request Routing:3μs

Active Connections:2.5M

DDoS Mitigation Rate:100%

The Skills

Certificates

CERT-ID: AWS-ASA-73921
VALIDATION: OK
ISSUER: Amazon Web Services
SCORE: 920/1000

Certificate 01

Supervised Machine Learning

An online non-credit course authorized by DeepLearning.AI and Stanford University and offered through Coursera.

Designation:Machine Learning

Validation Details:Completed

import tensorflow as tf

model = tf.keras.models.Sequential([
  tf.keras.layers.Dense(128, activation='relu'),
  tf.keras.layers.Dense(10)
])

Certificate 02

NPTEL

Introduction to Internet of Things

Score:90%

Top:5%

const UI = () => (
  <div className="grid">
    <PremiumAesthetic />
  </div>
);
export default UI;

Certificate 03

Machine Learning with Python

Advanced certification in React, UI/UX principles, and responsive web design.

Focus:Machine Learning

Projects completed:5

SELECT 
  user_id, 
  COUNT(*) 
FROM 
  analytics.events 
GROUP BY 
  user_id;

Certificate 04

Oracle Cloud Infrastructure 2025 .

Certified AI Foundations Associate

Focus:Cloud Infrastructure

time:25+ hours

contact

Build the future.

GitHub

Mail

Medium