I am a B.Tech Computer Science Engineering and Data Science student at Dwarkadas J. Sanghvi College of Engineering (CGPA: 9.17/10), passionate about building high-impact systems at the intersection of reinforcement learning and AI-driven automation, browser and system-level instrumentation and performance signals, and full-stack product engineering for real-world applications.

I enjoy solving complex engineering problems end-to-end — from architecture and infrastructure to model integration and production deployments.

Location: Mumbai, India  |  Email: jenithjain09@gmail.com  |  Phone: +91-9022823830

Dec 2025 – Jan 2026 | Remote

• Architected a multi-container Docker setup (Ray, Android emulator, task containers) to run RL experiments on SWE-bench tasks.
• Designed a multi-signal reward pipeline combining SSIM, accessibility tree diffs, Chromium performance metrics, patch similarity, and unit test outcomes.
• Collected browser-level rewards from a real Android x86_64 emulator running a custom Chromium APK via CDP (GPU memory, paint time, LCP, CLS).

May 2025 – Jun 2025 | Remote

• Built a cryptography library in C implementing AES and RSA primitives for secure embedded systems.
• Collaborated on MCU integration tests and optimized for memory-constrained firmware environments.

Jun 2025 – Jul 2025 | Remote

• Worked on deep learning pipelines for movie and video dubbing, including dataset annotation and preprocessing.
• Implemented model workflows for automatic dubbing and voice conversion using Python.

• Built RL infrastructure for automated bug fixing (SWE-bench) and frontend generation (Design2Code).
• Ran experiments across 484 SWE-bench + 484 Design2Code tasks.
• Developed a fully Dockerized system with Ray workers, Android emulators, and task containers.
• Integrated 5 reward channels: visual similarity, semantic DOM diff, Chromium performance, patch similarity, and test execution.

• Built a cyber-defense platform to detect phishing, malicious URLs, prompt injection, and behavioral anomalies.
• Implemented real-time deepfake detection pipeline using MTCNN + EfficientNet with artifact analysis via Flask.
• Developed a Manifest V3 Chrome extension to intercept navigation and email contexts and render threat overlays.
• Designed a Neo4j knowledge graph linking users, domains, IPs, and DDoS indicators.
• Implemented 5-layer PII anonymization (DROP/HASH/BUCKET/KEEP/ADD) and Merkle-tree audit trails.

• Architected a production-grade multi-agent system with 7 specialized agents using Vercel AI SDK and Zod.
• Built Neo4j-backed persistent memory with entity extraction, relationship mapping, and cross-draft reasoning.
• Routed heavy graph operations to reasoning-focused agents and lightweight tasks to fast-inference agents, reducing LLM costs by 80%.
• Enforced AES-256-GCM per-user encryption at rest with request-time decryption and zero plaintext persistence.

• 3x Hackathon Winner, 11x Finalist
• Winner — SPIT-CSI National Level Hackathon (3000+ registrations)
• Finalist — Odoo Hackathon at IIT Gandhinagar
• 2x Problem Statement Winner — CODESHASTRA XI and XII
• 2x Winner (Maharashtra Round) — Bit N Build International Hackathon

• Software Engineering Internships
• AI and ML Engineering Roles
• Systems and Infrastructure Engineering Opportunities
• Full-Stack and Applied AI Product Building

If you are hiring or collaborating on ambitious projects, feel free to reach out.

Email: jenithjain09@gmail.com  |  LinkedIn: linkedin.com/in/jenithjain