Projects

I am quite active on Github. Below, is a list of some of my selected projects.
Estimation of Remaining Useful Life of Bearings using ML
Supervised by Prof. S.P. Harsha
  • Built a predictive-maintenance framework estimating the Remaining Useful Life of high-speed bearings.
  • Collected vibration signals from an experimental test rig and denoised them via wavelet thresholding.
  • Extracted 34 multi-domain features and filtered them using S-weighted and Pearson correlation scoring.
  • Designed a self-attention-augmented BiLSTM to model temporal degradation patterns in vibration data.
  • Tuned hyperparameters via Bayesian optimisation, selecting hidden units, dropout, and learning rate.
  • Achieved R² of 0.825 on combined data and 0.781 on the held-out bearing under leave-one-out testing.
Technologies involved: Python, PyTorch, BiLSTM with Self-Attention, Bayesian Optimisation
Simulation of Human-Induced Load Patterns using ML
Supervised by Prof. Anil Kumar
  • Developed a GAN-based model generating human-induced load patterns at 98% similarity to real data.
  • Processed 658+ experimental samples across walking, jumping, and bouncing loads at 1.5 to 3.5 Hz.
  • Built a Forecasting+LSTM hybrid model and studied ARIMA and Prophet for time-series load synthesis.
  • Validated synthetic loads via PSD replication and statistical equivalence in structural response.
Technologies involved: Python, GANs, LSTM, ARIMA, Prophet, Time-Series Modelling
Design and Analysis of Inflatable Decoy
Supervised by Prof. S.H. Upadhyay
  • Designed a 1/12-scaled inflatable decoy of a missile launcher using a PVC-coated polyester fabric.
  • Compared different geometries via ABAQUS FEA, choosing the pipe structure for bulge control.
  • Simulated inflation–deflation cycles in ABAQUS using the fabric material model and fluid cavities.
  • Built a physical prototype using 3D-printed joints and tested radar and thermal-IR signature samples.
Technologies involved: ABAQUS, FEA, SolidWorks, 3D Printing, Radar/IR Signature Engineering
Programmable Soft Biomimetic Crawler
  • Designed a soft biomimetic crawler inspired by worm and caterpillar peristaltic-undulatory motion, aimed at navigating confined and unstructured terrains where conventional rigid robots struggle.
  • Modelled the body geometry in SolidWorks with a focus on flexibility and miniaturisation.
  • Ran ANSYS FEA to simulate deformation under actuation and validate the soft body's crawling gait.
  • Engineered programmable magnetisation patterns inside the body so that an external magnetic field could drive controlled, untethered crawling motion.
Technologies involved: SolidWorks, ANSYS FEA, Soft Robotics, Magnetic Actuation, Biomimetic Design
CAE Design and Structural Analysis of Roll Cage
Supervised by Prof. Indra Vir Singh
  • Reviewed roll cage design standards and prior reports to anchor the project's safety and load brief.
  • Built and iteratively optimised welded CAD models of the roll cage to maximise structural strength.
  • Ran ANSYS impact simulations to identify deformation modes and address structural safety concerns.
Technologies involved: SolidWorks, ANSYS, FEA, Structural Analysis
Vibration Control of Supported Beams
Supervised by Prof. M.M. Joglekar
  • Modelled a supported beam with and without an attached spring-mass-damper subsystem at the midspan.
  • Ran MATLAB simulations to compare dynamic responses and quantify the SMD's vibration suppression.
Technologies involved: MATLAB, Structural Dynamics, Spring-Mass-Damper Modelling
Code Development of 1-D Fanno Flow
Supervised by Prof. Ankit Bansal
  • Built a MATLAB tool simulating 1-D Fanno flow with shock waves and frictional gas-dynamics effects.
  • Implemented governing equations, validated against analytical results, and tuned numerical stability.
Technologies involved: MATLAB, Compressible Flow, Numerical Methods
image-denoiser Star
  • Designed an autoencoder-based image denoiser for low-light photos, reaching a PSNR of 28.256 dB.
  • Applied RIDNet to improve visibility and detail retention, raising PSNR to 30.579 dB on the same set.
Technologies involved: Python, PyTorch, Autoencoders, RIDNet, Computer Vision View Source
stock-sentiment Star
  • Built an NLP-driven sentiment model to predict stock-price movements from financial news headlines.
  • Back-tested the trading strategy on historical data, yielding 17.12% returns and a 1.34 Sharpe ratio.
Technologies involved: Python, NLP, Sentiment Analysis, Back-testing View Source
  • Built a backend service that periodically checks website health and tracks uptime status over time.
  • Set up background cron jobs to automatically schedule and run health checks every hour.
  • Created REST APIs for managing monitored sites and viewing their current uptime status.
  • Built a Slack integration to instantly alert users whenever a monitored site goes down.
Technologies involved: Go, Encore, PostgreSQL, GORM, Pub/Sub, Slack API. View Source
  • Built a BitTorrent client that allows downloads using a single-file torrent or magnet links
  • Implemented p2p protocol with tracker communction, piece checks, metadata exchange
Technologies involved: Java, BitTorrent protocol, TCP Sockets, HTTP Client View Source
  • Built a caching reverse proxy CLI tool which reduces server hits and request latency.
  • Implemented a thread safe concurrent disk based cache storage system.
Technologies involved: Go, net/http, SHA-256 hashing View Source
  • Built a git client that clones repositories using the git smart http protocol.
  • Handles git objects and writes files directly, eliminating the need for ZIP extraction.
Technologies involved: Java, Git Smart HTTP Protocol, SHA-1 hashing View Source
  • Built in-memory Redis-style server and CLI client in Go which supports redis core commands.
  • Implemented thread-safe storage for strings (with expiry), lists (with blockings operations), sorted sets.
  • Supports multiple concurrent clients and load an existing RDB file for persistence.
Technologies involved: Go, TCP/IP Networking, RESP Protocol, Concurrency View Source
  • built a custom load balancer using round-robin scheduling to distribute traffic across servers.
  • Added automatic health checks and reroute requests to healthy servers.
Technologies involved: Go, net/http, ReverseProxy View Source
  • bytehell is a UNIX-like command-line shell used to execute built-in Linux commands.
Technologies involved: C, POSIX system calls (fork, exec, wait), Unix/Linux View Source