atharva = {
"role": "AI/ML Engineer & Founder",
"degree": "Master of Applied AI — Deakin University (2024–2026)",
"currently": ["AI Developer @ InnovAIte (Capstone)",
"Software Engineer @ Deakin Competitive Robotics"],
"shipped": "ClauseCheck AU — live AI SaaS analysing Australian construction contracts",
"interests": ["Computer Vision", "LLM Agents", "MLOps", "ROS 2", "Full-Stack AI"],
"ask_me": "computer vision · LLM agents · MLOps · FastAPI · ROS 2",
}ClauseCheck AU — Live AI SaaS
Sole founder. Analyses Australian construction contracts (AS4000, GC21, ABIC) and flags risky clauses in plain English in under 2 minutes.
| Layer | Stack |
|---|---|
| Frontend | TanStack Start · TypeScript · Cloudflare Workers (SSR) |
| Backend | Python · REST API · Render · Supabase PostgreSQL + RLS |
| AI | Claude Opus 4.6 + Haiku · async pipeline · Supabase Realtime |
| Payments & Auth | Stripe AUD · Google OAuth · Deno Edge Functions · 4 pricing tiers |
| CI/CD | GitHub Actions · 12 AI eval pytest tests (all green) · auto-deploy on push |
Results: 21-finding report on real AS4000 contract · state-specific SoP Act rules for all 8 AU states · client-side PDF validation saves API credits
Represented Deakin University — 12th place out of 17 teams · 390+ students across 5 countries
![]() |
|
| Receiving the award | Team with Borealis rover |
Team with Borealis rover
| Project | What I'm building | Stack |
|---|---|---|
| FitTrack | Voice-first fitness & calorie tracker — Alexa logs workouts/meals via Lambda into a Render backend with a React dashboard | Alexa · AWS Lambda · Python · React |
| DermaGuard AI | Full-stack skin cancer screening platform — AI image analysis, PDF reports, hospital discovery | PyTorch EfficientNet-B4 · Python · React · Google Places API |
| Project | Highlights |
|---|---|
| StudyMate AI | Agentic Socratic tutor · 3-level adaptive teaching · repo-aware · deployed on Render · submitted to Sophiie AI Hackathon ($5K prize, 200+ devs) |
| Mine Safety RAG Copilot | NLP Q&A over 10+ safety manuals · 1,000+ chunk RAG pipeline · 75% query time reduction · YOLO PPE detection across 3 camera feeds |
| Azure ML Predictive Maintenance | SDK v2 pipeline · custom feature engineering · deployed to managed online endpoint for real-time inference |
AI & ML
Web & Backend
DevOps & Cloud
Robotics


