name: Ravi Kumar Keshari
education:
institute: Indian Institute of Technology Patna
program: B.S. Computer Science & Data Analytics
timeline: 2024 – 2028
core_expertise:
- Scalable Backend Infrastructure
- Generative AI & Diffusion Models
- Retrieval-Augmented Generation (RAG)
- Full Stack Product Engineering
- System Design for High Concurrency
philosophy: >
Ship systems that scale, measure everything,
and refactor without mercy. |
|
| Domain | Proficiency | Details |
|---|---|---|
| Full Stack Engineering | ██████████ | Next.js, NestJS, Flutter — end-to-end multi-surface architecture |
| Backend & System Design | ██████████ | Distributed systems, Redis queues, REST/WebSockets, high concurrency |
| Generative AI & RAG | █████████░ | Multimodal RAG pipelines, Diffusion Models, ViT, pgvector |
| Data Science | █████████░ | Pandas, NumPy, scikit-learn — analysis & applied ML |
| DevOps & Cloud | ████████░░ | Docker, CI/CD pipelines, AWS, Edge deployment |
| DSA & Algorithms | █████████░ | Daily problem solving, optimized system design for hackathons |
Core Languages
Backend & Databases
AI & Data Engineering
Frontend & Mobile
Cloud, DevOps & Tooling
|
Healthcare Networking Ecosystem
Architected an MVP for 100,000+ users. Engineered a decoupled backend using Redis queues for asynchronous job processing, ensuring non-blocking real-time mentor matching and notifications.
|
AI-Powered EdTech SaaS
Built a high-throughput multimodal RAG tutor trained on NCERT data. Parsed, chunked, and embedded textbooks using |
|
AI Mock Interview Simulator
Designed a full system architecture with a real-time low-latency voice pipeline and adaptive technical question generation for interview simulation.
|
|
| Recognition | Details |
|---|---|
| 🥉 HackNtech 3.0 | Secured 3rd Position overall in the hackathon, competing against top teams. |
| 📚 Leazy Learn | Successfully shipped a production AI EdTech SaaS with a multimodal RAG tutor serving real students. |
learning:
- Advanced Distributed Systems (Paxos/Raft architectures)
- Edge AI & On-Device Inference optimization
- Advanced Data Structures & Algorithms
building:
- DocMateX — Enterprise-grade healthcare ecosystem
- Leazy Learn — Agentic AI Tutor workflows
exploring:
- AI Safety, Alignment, and Agentic Workflows
- Multi-Sensor Satellite Data Fusion
open_to:
- SDE Internships (Backend / Fullstack)
- Research Collaborations in Applied AI
- Open Source Contributions
- Hackathons & System Design Competitions|
|
|
|
|
|
"Scalability is not an afterthought; it is an architectural prerequisite."