Treballador / a familiar per a SAD Mollet

Descripción del trabajo

Founding Software Engineer

Expected outcomes:

1. Design and implement production-grade AI systems with focus on LLMs and autonomous agents.
2. Develop optimized RAG systems and embedding pipelines.
3. Build robust Python applications emphasizing async programming and performance.
4. Deploy, monitor and maintain AI systems in production.
5. Establish a scalable and maintainable technical foundation, balancing speed of delivery with long-term scalability.
6. Champion the use of AI, staying at the forefront of Agentic AI advancements, and apply these learnings to deliver value-driven solutions.
7. Be part of building our performance-driven culture that values transparency, simplicity, and rapid iteration.
8. Foster a culture of learning and innovation, enabling the team to adapt to the rapidly evolving AI landscape.

All the desired skills we are looking for:

1. You have experience building in a fast-iteration environment - ideally with start-up experience.
2. You solve problems quickly and enjoy delivering outcomes, not just fancy technology.
3. You love learning and being part of a high-performing team.
4. Data-first - you love thinking on a data-first mindset and its possibility to create better products, experiences and business models down the line.
5. You’ve been deploying agent architectures / LLMs in production and at scale.

On a technical note, our stack below:

1. Python expertise.
2. Async programming, API development, real-time ASGI, Django / Channels.
3. Testing, logging, monitoring, performance optimization.
4. Production-grade application development.
5. Knowledge of data storage solutions (both SQL and NoSQL).
6. Async message queues (Celery RabbitMQ / Redis).
7. Experience with high-throughput, low-latency systems.
8. LLMs usage.
9. Strong understanding of LLMs, prompt engineering, chaining, caching and model fine-tuning.
10. Experience with both open-source and closed-source LLMs.
11. Experience with Retrieval-Augmented Generation (RAG), embedding optimization, chunking strategies, vector databases (ChromaDB, Pinecone).
12. Experience with LLM frameworks (Llamaindex / Langchain).
13. Model evaluation metrics and performance benchmarks.
14. LLM guardrails and security best practices.
15. LLM cost-optimisation strategies.