# Nabh Patodi

> AI-ML and Backend Engineer based in Indore, India. Available globally for freelance projects, consulting, and full-time engineering roles. Specialises in production LLM features, RAG systems, multi-agent platforms, MLOps, and scalable backend architecture.

## Profile

- Name: Nabh Patodi
- Role: Generative AI / AI-ML / Backend Engineer / Software Development Engineer
- Location: Indore, Madhya Pradesh, India
- Email: nabhpatodi1005@gmail.com
- Phone: +91 76940 72747
- Education: B.Tech Computer Science & Engineering, SRM Institute of Science and Technology, Kattankulathur (Aug 2022 – May 2026, CGPA 8.92/10.0).
- Availability: Open to freelance, consulting, and full-time engineering roles globally.

## What I build

Core build tracks I typically deliver for product teams:

1. AI Product Features — From LLM prototypes to reliable production endpoints. RAG pipelines, agent workflows, guardrails + evals, prompt optimisation, context engineering.
2. Backend Core — High-throughput APIs and resilient service architecture. Service design, async queues, caching, indexing, auth and multi-tenant patterns.
3. MLOps + Reliability — Model lifecycle and operational safety from day one. Retraining pipelines, observability, CI/CD, experiment tracking, drift detection + alerting.

## Services

- AI/ML Engineering — LLM features, RAG systems, prompt pipelines, evaluation frameworks, and production-ready AI workflows.
- Backend Architecture — scalable API design, domain modelling, auth patterns, service boundaries, and maintainable backend systems.
- MLOps and Model Serving — model versioning, deployment automation, inference optimisation, and reliable serving for production ML systems.
- Data Pipelines and Integrations — batch and streaming pipelines, ETL orchestration, and backend integrations that keep production data moving.
- Observability and Monitoring — telemetry, alerts, dashboards, and SLO tracking across APIs, AI services, and ML workloads.

## Experience

### Software Development & AI Engineer — Comoto Holdings (Jul 2025 – Present, Remote/USA)

- Built and deployed a microservice-based multi-agent platform using LangChain/LangGraph, leveraging PostgreSQL and LLMs from various providers, with <7s E2E latency.
- Designed a real-time RAG pipeline and custom data connectors to integrate multiple internal data sources, enabling agents to retrieve and reason over enterprise knowledge securely and reliably.
- Developed an analytics dashboard for observability with prompt and agent management, RBAC, Google SSO integration; and deployed the agents and dashboard across on-prem datacentre and GCP.

Technologies: Python, LangChain, LangGraph, PostgreSQL, FastAPI, React.js, GCP, Google SSO, RBAC.

### R&D Intern — Samsung R&D Institute India (SRI-B), Bengaluru (Jan 2024 – Sept 2024)

- Trained and evaluated classification models on sensor/health signals; compared against baselines and iterated on features and validation.
- Prototyped secure IoT data flow for on-device readings → gateway → analytics; contributed to an MVP integrating ML-driven insights.

Technologies: Python, multiclass classification, sensor/health data, MLOps, IoT.

## Featured Projects

### Research-AI
- Summary: Full-stack deep-research platform for generating citation-grounded long-form research documents with multi-agent architecture.
- Highlights:
  - Graph-based workflows for parallel search, scraping, and citation-grounded research document generation.
  - Pluggable LLM providers, FastAPI backend with real-time RAG pipeline and multi-agent architecture.
  - Top performer on DeepResearch Bench test; beats OpenAI, Gemini, Perplexity and other deep-research services.
- Stack: FastAPI, React.js, Tailwind, Firebase, LangGraph, LangChain, OpenAI, Gemini.
- Source: https://github.com/nabhpatodi10/Research-AI

### Computer-Use Agent
- Summary: Multimodal desktop AI agent that translates natural-language goals into a perception → action loop controlling mouse and keyboard.
- Highlights:
  - Runs 100% locally or uses a hybrid approach with local and cloud-based LLMs to perform long and complex tasks.
  - Dual backends — Microsoft GUI-Actor vision and OmniParser + LLM parsing — behind a compact LangGraph agent.
- Stack: Python, LangGraph, GUI-Actor, OmniParser, Tkinter.
- Source: https://github.com/nabhpatodi10/Computer-Use-Agent

### Internal Multi-Agent Platform (at Comoto Holdings, USA)
- Summary: Built and deployed a microservice-based multi-agent platform with real-time RAG pipeline, multiple data source integrations, and sessioned chats.
- Highlights:
  - E2E latency <7s.
  - Custom analytics and monitoring dashboard.
  - RBAC and SSO integration for the dashboard and agents.
- Stack: GCP, LLMOps, PostgreSQL, FastAPI, LangChain, React.js.

### DigiBanker (Standard Chartered Hackathon 2025)
- Summary: Agentic AI Branch Manager that assists KYC/loan workflows supporting multimodal inputs.
- Highlights:
  - Document handling, basic eligibility checks, customer-facing AI agent, and multiple ML models for loan classification.
  - Prototype reported >90% classification accuracy with ~10s typical E2E latency.
- Stack: FastAPI, React.js, MongoDB, OpenCV, TensorFlow, Keras.
- Source: https://github.com/nabhpatodi10/DigiBanker

### Health Device Information Classification (at Samsung R&D Institute India — Bengaluru)
- Summary: Trained and evaluated 120+ classification models on sensor/health signals; compared against baselines and iterated on features and validation.
- Highlights:
  - ~85% average accuracy across multiple device types.
  - 10,000+ records processed.
  - Deployed models with a scheduled retraining pipeline.
- Stack: Python, multiclass classification, data processing, MLOps.

## Tech Stack (full)

- Languages: Python, Java, C/C++, JavaScript, TypeScript, R, SQL.
- AI/ML: PyTorch, TensorFlow, Keras, Scikit-learn, OpenCV, LangChain, LangGraph, CrewAI.
- Backend and data: FastAPI, Django, Flask, PostgreSQL, MongoDB, MySQL, Redis, Firebase.
- Full-stack: React.js, Next.js, Tailwind.
- Infrastructure and DevOps: Docker, Kubernetes, Helm, Git, CircleCI, CI/CD, Amazon Web Services (AWS), Google Cloud Platform (GCP), Microsoft Azure.

## Education

- SRM Institute of Science and Technology, Kattankulathur — B.Tech, Computer Science & Engineering (Aug 2022 – May 2026 expected). CGPA: 8.92 / 10.0.
- The Shishukunj International School, Indore — Class XII: 90.0% (2022). Class X: 93.8% (2020).

## Publications

- Nabh Patodi, Tanmay Agarwal, Madhumitha K. "User Identification Based on Health Device Readings." https://ssrn.com/abstract=5833942
- Nabh Patodi, Kushagra Saxena, Lokesh Jain, Shlok Balsara, Dr. Tyj Naga Malleswari. "Smart Security System with Voice Bot." International Journal of Innovation in Engineering Research & Management, 12(1), 54–60. https://journal.ijierm.co.in/index.php/ijierm/article/view/2633

## Frequently Asked Questions

**What kinds of clients do you usually work with?**
Mostly SaaS teams, AI startups, and product companies that need production-grade backend and ML execution without expanding permanent headcount.

**Can you work with our existing backend stack?**
Yes. I usually integrate with existing services and improve architecture incrementally, unless a clean rebuild is clearly the better long-term option.

**Do you also handle ML deployment and monitoring?**
Yes. I cover model serving, deployment workflows, metrics, logging, and alerting so ML features stay reliable after release.

**How quickly can we start?**
Usually within 3 to 7 days depending on scope and timeline. Architecture audits can often begin within 48 hours.

## Contact

- Email: nabhpatodi1005@gmail.com
- Phone: +91 76940 72747
- Book a 30-minute call: https://cal.com/nabhpatodi/30min
- LinkedIn: https://www.linkedin.com/in/nabhpatodi10
- GitHub: https://github.com/nabhpatodi10
- Upwork: https://www.upwork.com/freelancers/nabhpatodi

Portfolio: https://nabhpatodi.com/
Resume: https://resume.nabhpatodi.com/
Resume PDF: https://resume.nabhpatodi.com/Nabh_Patodi_Resume.pdf