Desarrollador/a Python

Senior Python Developer | Agentic AI & LLMs Join a leading technology company to build production-grade agentic AI that automates digital and broadcast media workflows for a global media client. Production delivery, not prototypes, in an Agile team.
What you'll do
- Architect an agentic harness from scratch and integrate agent frameworks where appropriate
- Build RESTful, streaming and async APIs, choosing between sync calls, queues, events and WebSockets
- Run large-scale batch pipelines with LLMs, VLMs and deep learning models over video and rich media
- Design and tune lexical and embedding-based search (relevance, ranking, indexing)
- Build eval datasets and run offline and online evals (rubric scoring, AI-as-judge, human review)
- Own production reliability and build front-end experiences in React/TypeScript
- Produce C4 architecture artifacts and document trade-offs on model selection, prompting and sampling
Must have
- 7+ years in Python; 5+ years shipping production software (APIs, queues, authN/authZ, observability)
- 6 to 12 months building agents: custom harness with state, RAG, sandboxing and streaming
- 1+ year evaluating AI systems: regression testing, AI-as-judge, precision/recall, statistical significance
- 2+ years architecting search at scale (OpenSearch/Elasticsearch)
- Foundation model knowledge beyond API calls: temperature, top-p, logprobs, model selection
- C4 model, Agile, BS in CS/EE or equivalent experience
- Upper-intermediate (B2) or advanced English
Nice to have
- Multimodal AI, video or broadcast production workflows
- Fine-tuning (LoRA, PEFT)
- 5+ years AWS, 1+ year Azure
- CI/CD, Docker, Kubernetes
- React, Angular, WebSockets, FFmpeg, OpenCV


Tech stack: Python (core), TypeScript/JavaScript for front-end AI: LLMs, VLMs, open-source models, RAG, OpenAI Agents SDK, LangGraph, Google ADK, SmolAgents Search: OpenSearch, Elasticsearch Front-end: React, TypeScript Cloud & DevOps: AWS, Azure, Docker, Kubernetes, CI/CD