Freelancer profile translated to English.
Back to original languageAbout Rasamison
I design and deploy AI systems in production — autonomous agents, RAG pipelines, backend APIs — with measurable results.
What I've delivered recently:
Real-time multi-agent system for a call center of 50 operators → -35% processing time, latency maintained < 200 ms
RAG pipeline on +10,000 legal texts → +40% accuracy vs classic search, FastAPI API < 300 ms
Application management deployed in production → -60% data collection time
What I can do for you:
Build an AI agent or chatbot connected to your data
Set up a RAG system on your internal documents
Create or optimize your backend APIs (FastAPI, NestJS, Express)
Automate your data workflows (scraping, ETL, pipelines)
Integrate OpenAI, Claude, Vertex AI, or open-source models into your applications
French
Native or bilingual
English
Fluent
Can work on-site
Paris (up to 50km)
Experience
- EtechAI EngineerFebruary 2025 - May 2026 (1 year and 3 months)Objective: assist operators in real-time — contextual response suggestion, automatic call summarization, and intelligent customer ticket escalation.● Architected and end-to-end managed the multi-agent system, reducing processing time by 35%.● Defined orchestration patterns with LangGraph (state graphs, specialized nodes) via gRPC and WebSocket, maintaining latency < 200 ms.● Designed the hybrid dense/sparse search strategy in Qdrant and LLM context caching via Redis.● Scalable deployment on GCP with Docker, zero downtime in production.● Conducted training workshop at the client's site for operational teams to adopt and use the system.
- Mission freelanceAI DeveloperMay 2025 - August 2025 (2 months)Tananarive, MadagascarDesign and implementation of a RAG system• Automatic scraping of legal texts with SeleniumBase• Preprocessing and chunking of the corpus, then generation of embeddings using OpenAI text-embedding-3-small• Indexing of vectors in Qdrant, hosted on an OVH server, with management of persistence and scaling• Deployment of a FastAPI API served by Uvicorn offering an endpoint for queries in hybrid search mode (dense + sparse) via LangChainTechnologies: LangChain, Qdrant, FastAPI, SeleniumBase, OpenAI, Docker
Recommendations
Be the first to recommend Rasamison
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 Computer ScienceNational School of Informatics (ENI), University of Fianarantsoa2024Master en informatique
- Bachelor in Computer ScienceNational School of Informatics (ENI), University of Fianarantsoa2022Licence en informatique