Who I Am
I'm a Computer Science student at MS Ramaiah Institute of Technology, Bengaluru, currently pursuing my B.E. in CS&E with a GPA of 9.17 / 10.
I like to build AI-driven systems that turn complex problems into practical solutions. My work focuses on deploying and scaling applications, building retrieval-augmented systems, and creating tools that bridge intelligent models with real-world use cases.
When I'm not coding, you'll find me organizing hackathons with ACM RIT, competing in coding competitions and hackathons across India, or exploring the latest in AI, Off-Screen I'm building pokemon teams.
Bengaluru, India
saksham.jadav@gmail.com
+91-7678369133

Education
B.E. Computer Science & Engineering
MS Ramaiah Institute of Technology
Oct 2023 – Oct 2027
12th — PCMCS
97%
Where I've Worked
Software Engineer Intern
Unisys- >Engineered an interactive graph caller tool to visualize legacy program flows, improving maintainability and reducing architecture understanding time by 30%.
- >Implemented secure HTTPS communication using SSL certificates and deployed with FastAPI + Nginx on RHEL servers.
What I've Built
Money Muling Detection Engine
Graph-Based Fraud Detection Backend- >Built a deterministic fraud-detection backend using graph algorithms including depth-limited DFS cycle detection, sliding-window smurfing analysis, and shell-network traversal.
- >Engineered structural feature extraction and rule-based risk scoring pipelines capable of analysing 10k+ transactions with explainable outputs.
System Platform
Distributed Job Processing Architecture- >Designed a distributed Python backend with asynchronous FastAPI services, Redis-backed job queues, worker processes, and scalable API orchestration.
- >Implemented caching strategies, token-bucket rate limiting, and background execution pipelines to improve throughput and system resilience.
VoiceCare
Parkinson's Risk Detection- >Developed a speech-processing ML pipeline achieving 91% accuracy across 1,200+ recordings using HuBERT-based feature extraction workflows.
- >Implemented explainable AI analysis with SHAP and structured inference pipelines for interpretable neurological risk assessment.
VocalCanvas
Voice-First AI Creation Platform- >Built a cross-platform voice-first AI creation tool enabling users to generate and interact with content through natural speech input.
- >Integrated Firebase backend with real-time speech-to-text processing for seamless voice-driven creative workflows.
ReNova
GenAI Waste Intelligence Platform- >Built a multimodal GenAI platform using CLIP vision, Whisper voice, and Llama-powered Dual-RAG (Global + Personal) for hyper-personalized recycling guidance.
- >Engineered backend services with FastAPI handling image classification, voice transcription, geospatial recycler ranking via OSRM, and token-based gamification.
SupportRAG
Retrieval Augmented Support Platform- >Designed a scalable backend service processing 15,580+ support documents using FAISS indexing, asynchronous FastAPI APIs, and modular retrieval workflows.
- >Deployed production-style REST services on Microsoft Azure App Service with sub-250ms response latency.
Whisper Walls
Location-Based Social Platform- >Developed a geospatial backend using MongoDB indexing to enable proximity-based content discovery and anonymous session-based interactions.
- >Designed RESTful APIs supporting real-time updates, discovery tracking, and scalable location-query workflows.
My Toolkit
Languages
Backend & Systems
Databases & Storage
Tools & Frameworks
AI & Data
Relevant Coursework
Competitions & Awards
Get In Touch
I'm always open to discussing new opportunities, interesting projects, or just having a conversation about backend systems and distributed architectures.