# Raphael Motta

Raphael Motta is a remote worker on Himalayas, Forward Deployed AI Engineer at Factored, based in Brazil.

Source: https://himalayas.app/@raphaelmotta

```text
Forward Deployed AI Engineer at Factored who improved enterprise RAG accuracy to 98% while cutting latency and inference costs.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | Brazil |
| Authorized to work in | Brazil |
| Salary expectations | 72k+ USD |
| Job categories | AI Engineer |
| Skills | Multi Agent Systems, RAG, Agentic Systems, AI Agents, Model Context Protocol (MCP), Graph RAG, Knowledge Graphs, Neo4j, Context Engineering, Semantic Search, Hybrid Retrieval, Model Evaluation, LLMOps, Human in the Loop AI, Distributed Systems |
| LinkedIn | https://linkedin.com/in/raphael-motta-646b53227 |

## What I'm looking for

```text
I’m available for fully remote B2B contract work, either part-time or full-time, including fixed-scope RAG audits and build sprints. I’m based in Brazil and not available for CLT, hybrid, or on-site roles.
```

## About

```text
At Factored, I optimized a production RAG pipeline for a major US enterprise, reaching 98% retrieval accuracy while reducing time to first token by 60% and inference cost by 30%. I validated the changes with offline retrieval evaluations.

I also re-architected a semantic compiler and agent orchestration workflow that turns software repositories into Neo4j knowledge graphs for persistent coding-agent context. The work improved context retrieval accuracy by 20% and reduced pipeline costs by 30%.

At All Integra, I built agentic RAG workflows for e-commerce product integration and a FastMCP service for product updates. These projects reduced onboarding errors by 50%, increased successful top-product delivery by 40%, and opened an AI-native channel that added 30% in revenue in its first three months.
```

## Experience

### Forward Deployed AI Engineer (current)

[Factored](https://himalayas.app/companies/factored). Apr 2026 - Present

```text
- Optimized a production RAG pipeline for a major US enterprise, replacing always-on reranking with hierarchical retrieval, HyDE, metadata filtering, and conditional reranking; validated with offline retrieval evaluations. Result: 98% retrieval accuracy, 60% faster time to first token, 30% lower inference cost.
- Re-architected a semantic compiler and agent orchestration workflow that turns softwa
```

### Forward Deployed AI Engineer

[BTG Pactual](https://himalayas.app/companies/btg-pactual). Sep 2025 - Apr 2026

```text
Fixed-scope contract engagement.

- Modeled financial documents, business rules, and domain relationships as Neo4j-backed ontologies and knowledge graphs for GenAI workflows.

- Built retrieval and tool-enabled agents grounded in graph context to support reasoning across trading and tax workflows, with automated evaluations of answer accuracy and grounding before release and monitoring in Amazon C
```

### Forward Deployed AI Engineer

Umbrage. Jun 2025 - Sep 2025

```text
Fixed-scope contract engagement.

- Built and deployed a multi-agent HR onboarding platform that organized policies and procedures in Neo4j; combined LangGraph, PydanticAI, Qdrant, Bedrock, and MCP integrations with Slack, Notion, and Outlook, deployed on EKS to automate policy assistance and equipment provisioning.

- Built a multi-tenant .NET and AWS platform for Weatherford that let clients lau
```

### AI Engineer

All Integra. Apr 2023 - Jun 2025

```text
- Built agentic RAG workflows with Neo4j, Qdrant, Cognee, and LangGraph to classify and integrate products across e-commerce platforms, using evaluation sets to track classification accuracy and human-in-the-loop product validation; reduced onboarding errors by 50% and increased successful top-product delivery by 40%.

- Mapped legacy codebases into knowledge graphs to give coding agents architect
```

### Software Engineer

All Integra. Apr 2021 - Apr 2023

```text
- Architected a distributed integration platform supporting 15+ marketplaces, including Amazon, Shopee, Mercado Libre, and TikTok Shop, processing 2M+ third-party API requests daily. Designed event-driven workflows with RabbitMQ and Redis to handle asynchronous processing, rate limits, retries, and external API failures, operating across Kubernetes-based production environments with ~99% availabil
```

## Education

### Universidade Veiga de Almeida

Bachelor of Engineering - BE, Computer Engineering. 2022 - 2027

### OpenAI

API Deployment Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### OpenAI

ChatGPT Deployment Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### OpenAI

ChatGPT Solutions Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### OpenAI

Codex Deployment Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### OpenAI

Codex Solutions Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### OpenAI

OpenAI Consultative Solutions Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### OpenAI

OpenAI Cyber Deployment Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### OpenAI

OpenAI Technical Practitioner

```text
Oct 2026 · Expires Oct 2027
```

### Udemy

Curso de JavaScript e Typescript do básico ao avançado

```text
Mar 2023
```

## Tech stack

- Microsoft Azure
- Amazon CloudWatch
- Amazon S3
- Kubernetes
- Amazon EKS
- GitHub Actions
- Next.js
- Databricks
- Neo4j
- Redis
- React
- Python
- C#
- Go
- RabbitMQ
- Amazon SQS
- FastAPI
- Amazon SNS
- Kibana
- SQLAlchemy
- Linux
- Datadog
- Gemini
- AWS Lambda
- Amazon RDS
- TypeScript
- Docker
- Amazon Web Services (AWS)
- Amazon SageMaker
- Clickhouse
- Qdrant
- Pinecone
- CrewAI
- DeepEval
- Langfuse
- LangGraph
- LangSmith
- PydanticAI
- Model Context Protocol (MCP)
- Mastra
- GraphRAG
- FalkorDB
- Agno
- Retrieval-Augmented Generation (RAG)
- Apache Kafka
- Claude
- REST APIs
- Amazon ECS

To contact Raphael, message them from this profile with a Himalayas employer account: [Hire on Himalayas](https://himalayas.app/hire).

## People also viewed

- [Ricardo Cunha](https://himalayas.app/@ricardocunha): Principal Architect, ZappyRide by J.D.Power
- [Matheus Silva](https://himalayas.app/@matheussilva1): Software Engineer, AI Platform & LLMOps, Housecall Pro
- [Henrique Camargo](https://himalayas.app/@henriquecamargo): GenAI Engineer (Agents & RAG), SPC Brasil
- [Ruan Chaves](https://himalayas.app/@ruanchaves): AI Engineer, Xseed Solutions
- [Caua User](https://himalayas.app/@cauauser): Engenheiro de IA Pleno, Podium Educação
- [ezequiel User](https://himalayas.app/@ezequieluser1): AI Product Manager | AI Solutions Architect | Agentic Engineering, Agrotis
- [Alan Cesar Gonçalves](https://himalayas.app/@alancesargonalves): Backend & AI Engineer, GFT Technologies
- [Bruno Silva](https://himalayas.app/@brunosilva5): AI Solutions Architect, Deloitte
- [Alex Felix](https://himalayas.app/@alexfelix): Principal AI Engineer, Flui Solutions
- [Anderson Zheng](https://himalayas.app/@andersonzheng): AI Engineer, EZOps Cloud
