M
madhav_murali's photo
Madhav Murali
From India 06:05 PM (GMT+05:30)
$25/hr or $80,000/yr

Active 1 hour ago


Member since Aug 2026

Share this profile:

Software Engineer

AI-native engineer
Available for hire
Years of experience
1+ years
Experience level
Junior
Available for
Full-time, Part-time, Contract, Freelance
Available from
12 Aug 2026

Full-stack software engineer with hands-on experience in Python, C++, and React, building AI/ML systems, RESTful APIs, and benchmarking optimization algorithms for resource-constrained UAV systems. Co-authored a published research paper on a two- phase UAV search framework, benchmarking five TSP solvers (Genetic Algorithm, PSO, Nearest Neighbor, Brute Force, 2-opt) and validating YOLOv8s computer vision-based survivor detection and depth-based obstacle avoidance on real drone flights. Currently building backend services for an enterprise Agentic AI/RAG platform at UBS, with growing depth in transformer-based LLM architectures, debugging, and performance optimization. B.Tech in Computer Science and Engineering from NIT Calicut.

Languages

Employment History

Software Engineer at UBS Current 2025 - Now
Built backend services (Python, .NET) for an enterprise Agentic AI platform, integrating Retrieval-Augmented Generation (RAG) with real-time market data and internal knowledge repositories to deliver explainable, context-aware financial insights. • Delivered portfolio optimization, multi-step reasoning, and Devil's Advocate analysis features for the AI platform; collaborated cross-functionally with QA, product, and business stakeholders within sprint timelines. • Participated in code reviews, debugging, root cause analysis, and performance optimization across services, contributing to reliable, production-grade software delivery. • Full stack engineer on the Graduate Talent Program (GTP), building enterprise applications for UBS Advantage Suite using .NET (C#), Python, TypeScript, and REST APIs across the SDLC in an Agile/Scrum environment. • Built E2E test automation frameworks (Cypress) and an automated CI/CD release pipeline (Azure DevOps/GitLab) requiring no manual intervention, cutting manual testing effort by 20%.
Co-Author at NIT Calicut 2024 - 2025
Co-authored “A Novel Two-Phase Approach for UAV-Based Survivor Identification in Disaster-Hit Areas,” published at the ICIoTCT International Conference — validated with real drone flights, not simulation. • Benchmarked five TSP solvers (Genetic Algorithm, PSO, Nearest Neighbor, Brute Force, 2-opt) on computational complexity, path efficiency, and reward performance for resource-constrained UAV path planning. • Designed a two-phase UAV trajectory planning and path optimization system: spiral reconnaissance for full-area coverage, then TSP-based waypoint clustering with A* real-time pathfinding for multi-drone task allocation. • Trained a YOLOv8s computer vision model on the C2A disaster dataset for real-time object detection and survivor identification, and used Depth Anything V2 for monocular depth estimation and obstacle avoidance; validated on real DJI Tello UAV flights.
Software Engineer Intern at Paygeon 2024 - 2024
Trained and deployed a machine learning model (XGBoost, Python, TensorFlow, scikit-learn) to match loan applicants to lenders using features including credit score, acceptance rate, income level, debt-to-income ratio, and employment history; built the full- stack web app end-to-end (React, Tailwind CSS, JavaScript, REST APIs) to serve it in a live, user-facing product, including data preprocessing, feature engineering, and model evaluation. • Built a full-featured admin dashboard (React, Firebase, Tailwind CSS) with secu

Education

Bachelors in Computer Science at National Institute of Technology,Calicut 2021 - 2025