"I explore systems — not just to understand them, but to make them observable, scalable, and intelligent."

• 📍 Chennai, India
• 🎓 Computer Science Engineer
• 💻 Full Stack Developer • Mobile App Developer • Cloud Engineer
• ⚡ Domains: AI Systems • Enterprise Platforms • Behavioral Tech

A production-grade Oracle Fusion monitoring system built for real-world enterprise environments

Over the past 9 months at Doyensys (Intern → Full-Time), I’ve been actively involved in designing and developing Fusion Insights, a system used to monitor and analyze large-scale Oracle Fusion environments for clients like Pokémon and Seaspan.

Enterprise systems often lack:

• Real-time visibility into background jobs
• Clear diagnostics for failures
• Centralized monitoring across integrations

• Developed a real-time ESS job monitoring system
  → Detects failed, long-running, and high-frequency jobs
  → Built using Python + APScheduler

• Engineered ATP database session monitoring
  → Tracks active sessions and performance bottlenecks

• Built OIC integration tracking modules
  → Enables visibility into Oracle Integration Cloud workflows

• Designed environment sizing analytics dashboard
  → Built ETL pipelines (BIP / OTBI → ATP Database)
  → Helps teams understand system capacity and scaling needs

• Improved invoice classification accuracy to 95%
• Achieved through:
  • Prompt engineering
  • Business rule integration
  • Edge-case optimization

• Migrating architecture: Streamlit (prototype) → React + FastAPI (production)

• Improved:

  • Performance
  • Scalability
  • Maintainability

• Leading development efforts
• Managing a team of 3 interns
• Handling Git-based workflows and integration
• Deploying and managing Linux-based environments

Understanding distributed systems across cloud providers

During my internship phase:

• Built unified monitoring across AWS, Azure, GCP, Oracle
• Reduced API latency by 80% using SQLite caching + schedulers
• Developed interactive dashboards (React + Nivo)
• Implemented AI summaries (Azure OpenAI) for faster root cause analysis

Exploring systems beyond Earth

• Simulation-based environment design
• Game-engine driven architecture
• Combines creativity with technical modeling

➡️ https://github.com/lolpanda2004/M4RS

Designing systems for how humans actually work

• Dopamine-driven task execution model
• Reward prediction before task execution
• AI-based task categorization
• Focus-based progression system
• Built using Flutter + AI integration

Languages:
Python • JavaScript • Dart

Frontend & Mobile:
React • Flutter

Backend & Systems:
FastAPI • Node.js • Scheduling Systems

Cloud & Data:
AWS • Azure • GCP • Oracle Cloud API Oracle Fusion • Oracle OIC SQLite • ATP Database

AI / ML:
Prompt Engineering • TensorFlow • PyTorch • Azure OpenAI API • Groq API

• 💼 LinkedIn: https://www.linkedin.com/in/lathika-devanand/
• 🌐 Portfolio: https://lathikadevanand.dev/

"Great developers build features. Engineers build systems. Explorers understand why systems fail — and fix them."

This journey is ongoing.