What you’ll do - Build and fine-tune STT / TTS models and plug them into RAG pipelines with LangChain / LangGraph. - Ship low-latency, real-time streaming services on AWS / GCP / Azure using Docker + CI/CD. - Create reusable components, maintain vector stores, and monitor uptime & performance. Must-have - 2 + years hands-on with Whisper, Vosk, Tacotron, FastSpeech or similar. - Production LangChain experience and at least one deployed RAG workflow. - Cloud-native DevOps skills (serverless, containers, managed vector DBs). - **Fluent English (written & spoken).**: - Solid Python (asyncio) and basic TypeScript. Nice-to-have - Work with Colombian-Spanish speech corpora. - gRPC/RTP streaming, Kafka/Flink analytics, OSS contributions. Tipo de puesto: Tiempo completo Experiência: - AI Modelling: 2 años (Deseable) Idioma: - English (Obligatorio) Disponibilidad para viajar: - 25 % (Deseable)