Company Description :
MarsDevs is a remote-first, offshore digital product development company based in India.
We specialize in building MVPs, Mobile and Web Applications, SaaS Products, and vertical solutions for SMBs.
With a track record of delivering over 60 solutions in 10 countries and 12 different industry verticals, we focus on quality code, speed, and lasting technical :
- Build and scale Retrieval-Augmented Generation (RAG) systems.
- Fine-tune LLMs for specific use cases and improve performance.
- Develop multi-agent systems using LangChain / LangGraph or other equivalent tools.
- Work with vector databases and scale them using metadata strategies.
- Optimize and deploy transformer-based models.
- Handle large datasets and design efficient data pipelines.
- Address LLM limitations with prompt engineering and fallback logic.
- Collaborate with product, data, and engineering teams to ship AI solutions.
- Stay updated on new tools, research, and apply best-fit :
- Should have 4+ years of experience.
- Experience building large-scale RAG systems.
- Hands-on with LLM fine-tuning and transformers.
- Strong with vector DBs (Pinecone, FAISS, etc.) and metadata filtering.
- Proven experience with multi-agent AI systems.
- Solid Python skills, good grasp of ML fundamentals.
- Experience with LangChain, LangGraph.
- Comfortable deploying on AWS / GCP / Azure.
- Knows LLM limitations and practical workarounds.
- NLP and conversational AI :
- Knowledge of MLOps, model deployment workflows.
- Familiarity with Spark, Ray, or other distributed systems.
(ref : hirist.tech)