Skip to content
Available for AI Engineering roles

Roger Demello

Building systems that think, reason and ship.

AI Engineer focused on agents, retrieval, and real-world systems - reasoning, retrieving, automating, and scaling.

Download résumé

Currently exploring

  • ↳Autonomous Agents
  • ↳Retrieval Systems
  • ↳Machine Learning
◆ Nagpur, India
01About

Field Notes

2024

Fell for the math behind ML.

An electronics undergrad who got pulled into models, gradients, and messy real data.

2025

Built ML systems.

Sleep-disorder prediction at 87% accuracy - feature engineering, validation, the boring parts that matter.

2026

Started building agents.

Shipped 3+ LLM apps and autonomous agents to 200+ users at AI LifeBOT.

Now

Obsessed with making AI useful.

RAG that actually retrieves, agents that actually finish the task.

How I work

  • -Ship small, measure, iterate.
  • -Latency and reliability over leaderboard scores.
  • -Make retrieval honest; make agents finish.
  • -Document so the next person - or model - can pick it up.
02Projects

Case Studies

[01]

Executive Email Copilot

Problem
Executives spend hours triaging email.
Approach
Multi-agent architecture.RAG memory.Tool calling.
Result
20% reduction in irrelevant responses.
Stack
Python · FastAPI · LangChain · OpenAI API · Vector DBs · PostgreSQL
Source
[02]

DealSentry - Proposal Guard

Problem
Enterprises review proposals for compliance by hand - slow and inconsistent.
Approach
Rule engine + AI risk scoring.PDF / DOCX ingestion.Approval workflows + audit trails.CRM integrations.
Result
Less manual review, faster compliance sign-off.
Stack
React · TypeScript · Node.js · PostgreSQL · Azure OpenAI
Source
[03]

AI Marketing Agent

Problem
No way to engage every customer in real time across WhatsApp, Instagram, and Messenger.
Approach
AI agents for responses + follow-ups.Async multi-channel backend.
Result
Higher engagement efficiency, less manual work.
Stack
Python · MongoDB · OpenAI API · FastAPI · Messaging APIs
Source
03Stack

Toolkit

What I reach for - chosen because it ships, not because it's trendy.

Languages
PythonC++JavaSQL
AI & ML
LLMsRAGAgentsLangChainLlamaIndexOpenAI APIHugging FaceTensorFlowPyTorchScikit-learn
Backend
FastAPIFlaskREST APIsAsyncMicroservices
Data
PostgreSQLMongoDBVector DBsPandasNumPy
Cloud & Tools
AWSDockerGitCI/CDLinux
04Experience

Timeline

  1. 2025

    CFM, RCOEM

    Machine Learning Intern

    Sleep-disorder prediction · 87% accuracy · Python, Scikit-learn

  2. 2026

    AI LifeBOT

    AI Engineer Intern

    3+ LLM apps & agents · 200+ users · RAG, LangChain

  3. Next

    ?

    Open to AI Engineering roles

    Let's build something.

Selected highlights

  • -Deployed 3+ production LLM apps and autonomous agents to 200+ users.
  • -Cut inference latency through optimized retrieval and caching.
  • -Built a sleep-disorder ML model at 87% accuracy with rigorous validation.
  • -Designed AWS architecture - EC2, S3, IAM, Auto Scaling, Load Balancer.

Education

CGPA
  • B.Tech, Electronics & Communication8.9
  • Minor, AI & Machine Learning9.6

Credentials

  • AWS Certified Cloud Practitioner - 2025
  • 2nd Place, Bytesage AI National Hackathon
05Writing

Engineering Journal

Recent thoughts - short notes from building things.

  • May 2026

    Why most RAG systems fail.

    It's retrieval quality, not model size, that decides whether the answer is useful.

  • Apr 2026

    Latency matters more than model size.

    Users feel p95 latency. They never see your benchmark scores.

  • Apr 2026

    Building reliable agents.

    Guardrails, evals, and knowing when the agent should stop.

  • Mar 2026

    Evals are the real moat.

    If you can't measure it, you can't improve it - agents especially.

  • Feb 2026

    Prompt engineering is spec-writing.

    Be precise about the contract, not clever with the words.

06Contact

Get in touch

bash - contact

roger@demello:~$ contact

→ emailrogerdemello289@gmail.com
→ linkedinlinkedin.com/in/rogerdemello
→ githubgithub.com/rogerdemello

status:Open to AI Engineering roles