I'm a Forward Deployed Software Engineer and AI systems builder with 7+ years of experience designing and shipping production software across enterprise AI, backend systems, APIs, data platforms, and cloud infrastructure.

I've worked directly with Fortune 500 engineering teams to turn ambiguous business problems into production-grade systems — spanning LLM applications, RAG, AI agents, MCP integrations, document intelligence, event-driven architectures, and large-scale data pipelines.

I enjoy building things at the intersection of AI + infrastructure + real-world problems.

📍 Singapore
🌐 pranoy.dev
💼 LinkedIn
✉️ devpranoy (at) gmail (dot) com

• 🤖 AI Agents & LLM Applications — RAG, tool use, MCP integrations, agentic workflows and autonomous systems
• 🧠 AI Infrastructure — inference pipelines, vector retrieval, OCR, metadata pipelines and Human-in-the-Loop systems
• ⚙️ Backend & APIs — distributed systems, API integrations, event-driven architectures and streaming pipelines
• ☁️ Cloud Platforms — AWS, Azure, GCP, Docker and Kubernetes
• 🧩 Forward Deployed Engineering — translating complex customer problems into architectures, prototypes and production systems

Built and published an AI agent skill integrating Ultrahuman wearable data with OpenClaw via MCP, enabling agents to securely retrieve and summarize sleep, HRV, recovery, activity and physiological metrics.

2,800+ users

AI Agents MCP OpenClaw APIs Wearables

Built a web application that reconstructs 3D apartment layouts from floor plans, simulates Wi-Fi coverage and recommends optimal router placement.

Built for the Google DeepMind Gemini 3 Hackathon Singapore 2026.

Gemini 3D Wi-Fi Simulation AI

Built a Figma plugin for generating image variations and overlay text using Stable Diffusion.

1,900+ users

Stable Diffusion Figma Generative AI

2023 – 2026

Worked directly with Fortune 500 customers including Verizon and NASDAQ, owning technical discovery, architecture and delivery for enterprise software and AI initiatives.

Some of the systems I worked on include:

• Distributed AI document intelligence processing 500,000+ documents/week
• OCR + RAG + vector retrieval pipelines
• Human-in-the-Loop validation systems
• Enterprise API integrations
• Cloud-native data and metadata pipelines
• Solution architecture and rapid customer prototyping

2020 – 2023

Led a 10-engineer cross-functional AI team delivering production AI platforms across pharma and banking.

Worked on:

• ML inference pipelines
• Autonomous workflow engines
• Vector retrieval systems
• Streaming and batch data architectures
• Event-driven systems
• AI platforms serving 800,000+ active users

2019 – 2020

Built and scaled 4 startup products from inception to first paying customers.

Among them, architected a cross-border payments platform processing $20K+ weekly volume across 2,000+ active users.

Languages

Python JavaScript TypeScript Go C++

AI / Backend

LLMs RAG AI Agents MCP Agentic Workflows OCR Vector Search FastAPI Flask TensorFlow PyTorch

Cloud / Infrastructure

AWS Azure GCP Docker Kubernetes Jenkins Microservices

Data

PostgreSQL MySQL Streaming Pipelines Batch Processing Data Lakes Metadata Pipelines

Frontend

React Vue

I occasionally write and speak about AI, software engineering and technology.

• 🧠 Tech & Philosophy — Substack
• ⚡ REST API Performance: Python vs Go
• 🎥 Understanding Decentralized Finance
• 🗣️ Delivered a session on Deploying Custom LLMs using RAG to Production using Llama, Flask and Google Cloud for Google Developers Group Kozhikode

Birla Institute of Technology and Science, Pilani
Bachelor of Science in Computer Science

Udacity
Deep Learning Nanodegree
Product Management Nanodegree

Y Combinator
Startup School

I'm particularly interested in:

• Forward Deployed Engineering
• AI Agents
• Agent infrastructure
• LLM applications
• Enterprise AI
• Developer tools
• Applied AI problems

Feel free to reach out:

🐦 @devpranoy
💼 linkedin.com/in/devpranoy
🌐 pranoy.dev
📧 devpranoy (at) gmail (dot) com