Available for new roles · Qatar / India

Aaron
Sequeira

Full-Stack AI Engineer/

Full-Stack AI Engineer specialising in enterprise AI platforms, intelligent document automation and scalable software architecture. I build production-grade React/FastAPI systems around LLM and RAG pipelines — semantic and graph-based retrieval, vector databases, cloud-native APIs — and own the architecture decisions that keep them fast and maintainable.

4
Engineering roles
Baker Hughes → Mannai → Wood → Smarter.Codes
29+
Public repositories
AI, CV, full-stack, IoT
1,100+
Cameras under AI monitoring
National border system
110+
Students mentored
RAG & LangChain, as ACM Lead
01

What I actually build

Four things show up in nearly every project I ship. They're the reason a model gets past a notebook and into someone's workflow.

Enterprise AI architecture

Designing the AI layer of a production SaaS platform end to end — LLM and RAG pipelines, semantic search, pgvector, graph-based retrieval and enterprise knowledge indexing — and defending those choices in Engineering Design Reviews.

Intelligent document automation

Turning unstructured document estates into something queryable: extraction, chunking strategy, embedding choice and retrieval evaluation over thousands of real documents. Measured in retrieval time cut, not benchmark scores.

Computer vision in production

YOLO and TensorFlow models containerised and deployed against a live 1,100-camera security estate, tuned to cut false alerts rather than just win on a validation split.

The full delivery path

React and FastAPI services with PostgreSQL behind them, packaged in Docker, shipped on Azure through GitHub Actions, tested with Playwright and Pytest — plus the API design, auth architecture and engineering standards that keep them maintainable.

Education

B.Tech, Information Technology

Manipal University Jaipur · CGPA 7.93 · 2021 – 2025

OOP · Data Structures & Algorithm Analysis · Machine Learning · Web Technologies

Birla Public School, Doha · Class 12: 90.6% · Class 10: 89.6%

Currently learning

Agentic AI architectures, Kubernetes at scale, and evaluation tooling for retrieval quality.

Problem solving

1,600+ LeetCode rating · 5★ HackerRank

02

Experience

Enterprise SaaS, industrial inspection, national security infrastructure and oilfield services — four very different constraint sets.

Full-Stack AI Engineer

·Smarter.CodesCurrent

Mar 2026 — Present · Remote

Core contributor to AgileWriter, an enterprise AI document automation platform.

  • Building full-stack capability across AgileWriter with React, FastAPI, Python, TypeScript, PostgreSQL, Azure, Docker, Microsoft Graph, Redux Toolkit and TailwindCSS.
  • Designed the AI architecture: LLM/RAG pipelines, semantic search, pgvector, graph-based retrieval and enterprise knowledge indexing — enabling contextual reasoning across enterprise knowledge bases.
  • Led architectural decision-making through Engineering Design Reviews, API design, system architecture documentation and authentication architecture for a production-scale SaaS platform.
  • Set reusable engineering standards and drove performance optimisation across the platform.
  • Also shipping a high-security LLM assistant for banking FAQs and Playwright UI automation coverage.
ReactFastAPITypeScriptPostgreSQLpgvectorAzureDockerMicrosoft GraphRedux Toolkit

AI Engineer

·Wood Engineering Qatar

Nov 2025 — Mar 2026 · Doha, Qatar

ML for industrial asset-life optimisation, plus a RAG search layer over the internal document estate.

  • Trained TensorFlow / PyTorch / MATLAB models for CML (Condition Monitoring Location) asset-life optimisation on a 2,100+ sample dataset.
  • Built custom AI bots to extract relevant data fast, cutting processing time by 97%.
  • Delivered a RAG-based internal document search system (LangChain / LlamaIndex + Qdrant), cutting document retrieval time by 65%.
PyTorchTensorFlowMATLABLangChainLlamaIndexQdrant

Software Engineer · Lanco Engineering

·Mannai Corporation

Oct 2025 — Nov 2025 · Doha, Qatar

AI monitoring for a national Border Information System at camera-fleet scale.

  • Integrated AI monitoring across 1,100+ AXIS cameras via Milestone VMS and Ipsotek.
  • Deployed containerised YOLO / TensorFlow models for real-time violation analysis, cutting false alerts by 11%.
  • Scaled automated object detection into live security operations.
YOLOTensorFlowDockerMilestone VMSIpsotek

Software Engineering Intern

·Baker Hughes

Sep 2024 — Sep 2025 · Pune, India

Process automation and document intelligence for the Pune and Celle tech centres.

  • Architected and deployed a Technical Service Portal and Laboratory Management System for the Pune and Celle Tech Centers on Microsoft Power Platform.
  • Automated key workflows with real-time tracking and auditing, reducing service downtime by 70% and accelerating compliance validation by 40%.
  • Engineered an end-to-end LLM-powered document intelligence system on a RAG architecture, cutting manual data entry by 75%.
Power PlatformMicrosoft 365 DynamicsPythonReactAzure AI
03

Selected work

Grouped by what each project demonstrates rather than by language. Every card links to a working repository — several are live.

Roneira Document Intelligence

Enterprise document processing with Azure AI + GPT-4 and a full RAG pipeline.

FastAPIPostgreSQLRedisAzure AIDocker
1st place · LearnIT Hackathon

GenAI Browser Tool

Browser extension for summarisation, cited Q&A and 50+ language translation.

TypeScriptViteVitestPlaywrightChrome APIs

Roneira AI HIFI

Stock forecasting and portfolio platform that surfaces model disagreement.

Next.js 16React 19FastAPISupabaseUpstash Redis
3rd place · University Hackathon 8.0

AI Traffic Management System

Real-time vehicle detection driving adaptive signal timing.

YOLOv8FastAPIReactWebSocketDocker
Best prototype · ACM Mini Hacks 2.0

NexusHome IoT

Smart-home energy platform with predictive maintenance.

.NET 8SignalRML.NETInfluxDBRedis

CogniCore

Fully local RAG assistant — no API keys, no data leaving the machine.

LangChainPhi-3ChromaDBFastAPIStreamlit
04

Toolkit

Tools I've shipped production or near-production work with — not a list of everything I've read about.

Languages

PythonTypeScriptJavaScriptC#C++JavaMATLABSQLBash

AI & Machine Learning

LLMsRAG pipelinesLangChainLlamaIndexPrompt engineeringGraph-based retrievalPyTorchTensorFlowscikit-learnXGBoostYOLOOpenCVsentence-transformers

Backend

FastAPIDjangoFlaskNode.jsExpress.jsASP.NET CoreMicrosoft GraphCeleryWebSockets

Frontend

ReactNext.jsRedux ToolkitTanStack QueryZustandTailwind CSSFramer Motion

Data & Vector Stores

PostgreSQLpgvectorQdrantChromaDBMongoDBRedisInfluxDBSupabaseMySQL

Cloud & DevOps

AzureAWSDockerKubernetesGitHub ActionsJenkinsVercelRailwayPower Platform

Testing & Tooling

PlaywrightPytestVitestJIRAPostmanGitHub Actions
LangChainQdrantpgvectorFastAPINext.jsPyTorchYOLOAzureDockerMicrosoft GraphPostgreSQLRedisPlaywrightTensorFlowXGBoostReactLangChainQdrantpgvectorFastAPINext.jsPyTorchYOLOAzureDockerMicrosoft GraphPostgreSQLRedisPlaywrightTensorFlowXGBoostReact
05

Measured impact

Numbers reported from the work itself — each one tied to a specific system and a specific team.

97%

less document processing time

Wood Engineering · AI extraction bots

75%

less manual data entry

Baker Hughes · LLM document pipeline

70%

less service downtime

Baker Hughes · Technical Service Portal

65%

faster document retrieval

Wood Engineering · RAG search

40%

faster compliance validation

Baker Hughes · Power Platform

11%

fewer false security alerts

Mannai · containerised CV models

1,100+

cameras under AI monitoring

Mannai · Border Information System

110+

students mentored on RAG & LangChain

ACM Lead · Manipal University Jaipur

Recognition

1st placeLearnIT HackathonGenAI Browser Tool2024
3rd placeUniversity Hackathon 8.0AI Traffic Management2023
Best prototypeACM Mini Hacks 2.0NexusHome smart-home energy2023
Merit scholarshipSPDC — Qatar Govt & Indian EmbassyAcademic excellence2021–2025
GDSC & ACM LeadGoogle Developer Student Club & ACMLed technical ops for Hacktoberfest; mentored 110+ students on RAG & LangChain2022–2024

06 — Contact

Let’s build something that ships.

Open to AI engineering & full-stack roles — remote or Gulf-based. I’m most useful where enterprise AI systems have to survive contact with real users, real corpora and real compliance requirements.

+974 66900150 · +91 99877 26291