Freelancer profile translated to English.
Back to original languageAbout David
Does your team waste hours on manual tasks or do you need to integrate AI into your product without sacrificing your application's performance?
I help companies scale theirbackendandautomate complex processesusing autonomous agents andRAGarchitectures.
I'm David, anAI & Backend Engineer**. I specialize in building robust, production-ready **AI solutionsusingPython, Django**, and vector databases (like **Qdrantand Milvus), orchestrating intelligent workflows with **LangChain and LangGraph**.
Unlike other profiles that only experiment with prompts, my differential value is my solid foundation as an **E2E Backend Developer**. I not only integrate AI APIs, but I research, build the infrastructure, optimize response times (ultra-fast inferences), and ensure the architecture is **scalable and secure**.
How can I help you with your projects?
- **Multi-Agent Systems**: Automation of complex workflows (e.g., automatic email triage with cross-validation between LLMs).
- **Advanced RAG Architectures**: Chatbots and internal search engines connected to your own document base for accurate answers.
- **Backend Development**: Creation of high-performance REST APIs with Django and Django Ninja.
- **Model Optimization**: Reduction of latency times and operational costs.
I work on aproject basis or weekly hour blocksfor continuous maintenance. Write to me and let's analyze how AI can save your company time and money.
Spanish
Native or bilingual
English
Fluent
Remote only
Primarily works remotely
Experience
- DominicosAI engineer / Backend DeveloperSOFTWARE PUBLISHINGOctober 2025 - Today (10 months)Madrid, Spain
- Designed and implemented a hybrid search architecture (Dense + Sparse) with reranking from scratch, integrating it into multiple corporate projects and reducing search time to less than 0.3s on average (without reranker).
- Developed a high-performance asynchronous RESTful API with Django Ninja for query vectorization and interaction with Qdrant, achieving response times of 0.8s in production with the reranker model active.
- Optimized SOTA models (BAAI/bge-m3, Mmarco cross-encoders) on CPU using INT8 quantization and ONNX compilation with AVX-512 instructions, significantly reducing computational resource usage in non-GPU environments.
- Implemented request caching, achieving response times under 0.18s for the most frequently queried topics.
- Deployed and managed distributed infrastructure: Qdrant on AWS EC2 with strict network controls and inference API on dedicated OVH servers.
- Created an internal Django library/package to standardize the creation of collections, complex filters, and bulk data ingestion, accelerating the time-to-market for future integrations and increasing search matching in adopted projects.
- Implemented complete ecosystem observability with Zabbix (dashboards and alerts on Qdrant and server metrics).
Stack: Python, Django Ninja, Qdrant, HuggingFace, ONNX Runtime, AWS EC2, Docker, Zabbix. - Linkadia MediaAI engineer / Backend DeveloperPRESS AND MEDIAJune 2025 - October 2025 (4 months)Madrid, Spain
- Developed and deployed microservices with Django REST Framework on AWS EKS (auto-scaling with Karpenter, load balancing, TLS/SSL), managing environments with Docker and Docker Compose.
- Automated the integration of +5 different advertising APIs and data loading into legacy databases, saving the team an average of 2.5 hours daily in cost, CPA, and conversion reporting.
- Built a conversational AI system with GeminiAI and MilvusDB applying RAG to provide persistent memory to the agent, connected to Slack for automatic generation of insights on advertising campaigns.
- Implemented critical asynchronous tasks with Celery and RabbitMQ, ensuring resilience in high-volume data analysis processes.
- Wrote complete technical and API documentation with Sphinx and Swagger Docs; configured alerts and test flows with n8n.
Stack: Python, Django REST Framework, AWS EKS, Celery, RabbitMQ, GeminiAI, MilvusDB, Docker, n8n. - DominicosJunior Backend DeveloperSOFTWARE PUBLISHINGSeptember 2024 - June 2025 (9 months)Madrid, Spain
- Developed scalable automation processes in Python, Django, and Bash, integrating social media APIs (WhatsApp, Facebook, X/Twitter, YouTube, SoundCloud) to automate multimedia content publishing and management.
- Implemented web crawling scripts and a content recommendation engine for dominicos.org, increasing search matching and content relevance for end-users.
- Managed Linux servers and MariaDB (SQL) databases, including automated CSV report exports for internal stakeholders.
- Supported backend system maintenance for universities, NGOs, and publishers; occasionally collaborated on frontend with JavaScript and Bootstrap.
- Integrated AWS Bedrock for AI use cases; implemented Notion as a team ticketing and documentation tool.
Stack: Python, Django, Bash, SQL, MariaDB, AWS Bedrock, Linux, NumPy
Recommendations
Be the first to recommend David
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Master in Artificial Intelligence and BigDataDigitechFP2026Máster FP
- Cross-Platform Application Development (DAM)MEDAC2024Ciclo Formativo de Grado Superior
Certifications
- B2 First – Score 173.Cambridge English2023