NAViGO
Intelligent Navigation Engine
Computer vision based navigation system for autonomous vehicles with real-time obstacle detection.
HI, I'M
ANIRUDH SARVE
Computer Engineering Student | AI/ML Enthusiast | Builder
I build practical solutions using AI, software and design. Currently exploring how technology can solve real-world problems and create meaningful impact.
Currently Working On
AI powered products and real-world solutions
6+
Projects Completed
2+
Years of Experience
99%
Positive Feedback
1000+
Ideas to Build
FEATURED WORK
Projects that solve real problems
A collection of projects focused on AI, web development and innovative solutions.
1 of 6 • Swipe cards →
ProtoLink
Custom Network Protocol Lab
A custom TCP/IP protocol implementation to visualize packet transmission and acknowledgments.
Virtual Try-On
AI Powered Fashion Experience
Virtual try-on system for e-commerce using deep learning and computer vision.
Neural-Inertial Dead-Reckoning
15-State ESKF Sensor Fusion
Continuous-time sensor fusion fusing 6-DoF IMU and odometry with Ceres non-linear optimization for GNSS-denied robotics.
Multimodal Optical-SAR Fusion
Cross-Modal Satellite Reconstruction
Dual-stream cross-modal attention ResU-Net reconstructing cloud-occluded multispectral imagery using Sentinel-1 SAR.
VeriClaim-Agent
Autonomous Insurance Adjudication
Autonomous multi-agent workflow parsing clinical EHR documents, verifying policy clauses, and detecting fraud with audit trails.
ABOUT ME
Curious Builder and Problem Solver
I'm a Computer Engineering student with a strong interest in Artificial Intelligence, Machine Learning and software development. I enjoy building products, solving real-world problems and exploring new technologies.
AI/ML
Building intelligent solutions with real-world impact.
Full-Stack Development
Turning ideas into scalable web applications.
Entrepreneurship
Exploring business ideas and product development.
Leadership
Team collaboration, project management and growth mindset.
SKILLS & TOOLCHAIN
Technologies I Work With
Autonomous perception pipelines, distributed systems, deep learning architectures & modern full-stack web platforms
CAREER & TRACK RECORD
Professional Experience
Engineering multimodal AI perception pipelines, distributed backend architectures, and mission-critical autonomous systems.
NxtChaptr
Software Developer
Edge AI & Full-Stack Infrastructure • Remote / Mumbai
Nov 2024 – Aug 2025
⚡ 8m → <30s Listing Time
⚡ <3.5s In-Browser WASM Matting
⚡ Zero Cloud-GPU Fees
🔒 Atomic Redis Distributed Locks
- Multimodal AI Listing Automation: Integrated Google Gemini Vision APIs into merchant listing workflows, automating brand recognition, condition classification, and SEO metadata generation to slash listing turnaround from 8 minutes to under 30 seconds.
- Zero-Cost Client-Side Vision: Deployed on-device image segmentation using a U²-Net model compiled to WebAssembly (WASM), processing background cutouts in <3.5 seconds directly in-browser prior to Cloudinary cloud sync and eliminating cloud-GPU fees.
-
High-Concurrency State Architecture: Engineered resilient backend state management using Upstash Redis atomic distributed locks (
SET NX EX) and WebSockets to prevent race-condition double purchases on rare 1-of-1 catalog items during live pop-up drops.
Wroomie
Software Developer
Applied ML & Backend Engineering • Mumbai, India
Nov 2023 – Jan 2024
⚡ 94% Match Satisfaction
⚡ 35% Faster Query Latency
🗺️ 2dsphere Geospatial Search
🤖 Automated Media KYC Ingestion
- Behavioral Compatibility Clustering: Developed multi-attribute compatibility scoring algorithms clustering user lifestyle preferences across 4 behavioral vectors (sleep schedules, cleanliness, dietary rules, and study habits), reaching a 94% match satisfaction rate.
- Automated Media Processing: Built automated room verification and KYC asset validation pipelines using Cloudinary and OpenCV, standardizing room imagery and user profile assets across student housing zones.
- Geospatial Query Optimization: Architected high-throughput Express REST APIs backed by MongoDB 2dsphere indexes, reducing geographic profile and room discovery latency by 35% across high-density rental districts.
ISRO SIH & IRC
Lead Systems Architect & Robotics Engineer
Autonomous Navigation & Satellite Perception • National Contests
2024 – 2026
🏆 Winner: ISRO SIH (PS 26168)
⚡ 15-State ESKF Sensor Fusion
⚡ 0.42% Subterranean Drift
🛰️ 32.8 dB PSNR SAR Reconstruction
- Winning ISRO SIH Architecture: Authored the winning sensor fusion architecture for ISRO SIH PS 26168, developing a 15-state Error-State Kalman Filter (ESKF) with non-linear Ceres optimization bounding drift to 0.42% during complete GNSS blackouts.
- Multimodal Satellite Reconstruction: Engineered a dual-stream ResU-Net with cross-modal attention conditioning cloud-covered optical multispectral channels on Sentinel-1 SAR backscatter to reach 32.8 dB PSNR cloud penetration.
- Embedded Rover Navigation: Developed onboard kinematic telemetry, obstacle avoidance guidance, and sensor pipelines on embedded ROS2 compute platforms under the Indian Rover Challenge (IRC).