Shashi Bhushan
Open to work1 years experienceAI Engineer
M.Tech CSE student, Patent Holder, ML & Backend Engineer.
Standing Rank
Rank Not Available
Developer Badges
No badges earned yet
Skills & Technologies
28 skillsWork Experience
Patent: Coordinated Indoor Position Determination via Multi-Stage Signal Processing
VIT, Vellore, IN
- Delivered an indoor localization system using Wi-Fi RSSI fingerprinting and Bluetooth integration, achieving approx. 1.6 m positioning accuracy (MAE: 1.468 m, R2: 0.978) in complex environments.
- Deployed an ESP32-based autonomous Wi-Fi sniffer on a robotic car, automating radio-map generation and capturing 4,277 RSSI records across 209 access points- eliminating manual site surveys.
- Designed and optimized a Dynamic Stacking ensemble (KNN, Random Forest, XGBoost + Gradient Boosting meta-learner), cutting localization error by approx. 43% (RMSE: 3.124 to 1.787).
- Engineered a real-time data pipeline using MQTT, Raspberry Pi, and Python, transforming raw Wi-Fi scans into machine-learning-ready datasets for live location predictions.
- Implemented an A*-based indoor navigation module using NetworkX for optimal route generation and dynamic path recalculation from predicted user locations to destinations.
Projects
proofStack — Agentic AI Resume Fit & Competency Verification Platform
- Architected an asynchronous, multi-stage AI engine integrating Google Gemini Pro API via FastAPI, orchestrating a 7-stage concurrent extraction pipeline with strict latency optimizations to reduce overall processing time by 40%.
- Engineered a resilient background processing pipeline to decouple heavy PDF parsing and LLM inference; executed comprehensive database optimizations and an aggressive Redis caching strategy.
- Developed an interactive AI Interrogation system using RAG to conduct dynamic technical interviews on weak resume claims.
- Designed a hybrid, dual-tier authentication & authorization system integrating FirebaseAuth with PostgreSQL Row-Level Security (RLS).
- Integrated enterprise monetization via Cashfree payment gateway and automated webhook processing.
- Built a responsive SSR frontend using Next.js 16 (App Router), TypeScript, and Tailwind CSS.
AutoML Assistant
- Architected an end-to-end AutoML platform using Streamlit + Python to automate data upload, preprocessing, analysis, model training, and optimization workflows.
- Designed a modular ML backend supporting regression, classification, and time-series forecasting.
- Integrated LLM-powered model recommendation (RAG) using LangChain, Groq (Llama 3.1), and FAISS.
- Implemented SHAP-based explainability to make model predictions interpretable.
- Exposed trained models through a FastAPI inference API with persistent workspace/model management.
OncoFollow - LLM-Powered Clinical Workflow Platform
- Built an agentic workflow platform using Groq Cloud LLMs with role-specific prompt engineering.
- Designed a serverless API layer implementing LLM routing logic with role-specific prompt controls.
- Engineered a rule-based triage engine across 20 symptom categories and 4 scoring dimensions.
- Anchored all AI outputs in NCCN/ASCO clinical guidelines with citation-backed document summarization.
- Crafted real-time analytics dashboards (Recharts) with AI-powered trend analysis.
Leaderboard Standings
Leaderboard Position Pending
Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.
Assessment Highlights
Assessments Not Completed
Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.
AI Collaboration Score
AI Collaboration Score Pending
Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.
Role Compatibility Profile
Role Compatibility Analysis Pending
Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.
Achievements
1st Place, ARC Hackathon
1st Place, ARC Hackathon (ARC Lab, VIT) – Created a GUI-based AutoML tool that automatically benchmarks and recommends the best ML model for any given dataset.
About Details
Professional Bio
M.Tech Computer Science Engineering student and patent holder specializing in ML pipelines, distributed systems, and backend engineering. Experienced across the full stack, from edge hardware integration to LLM-powered SaaS platforms.
Integrated M.Tech in Computer Science Engineering in Computer Science Engineering
Vellore Institute of Technology, Vellore ( - 2027)
Junior College (Class XII)
SM Public School, Haridwar ( - 2021)
High School (Class X)
SM Public School, Haridwar ( - 2019)