Agentic AI Engineer

About PurpleMerit

We’re a next-gen AI startup building the future of intelligent systems, and we’re looking for a hardcore Agentic AI Engineer to help us push the limits.

If you know your way around building MCP, RAG pipelines, LLM-driven applications, and autonomous agent workflows, this role is built for you. You’ll architect, deploy, and fine-tune real AI systems that don’t just run… they think, adapt, and act.

If you love experimenting with cutting-edge models, breaking bottlenecks, and turning ideas into production-ready AI, come build with us. This is where fast-paced engineering meets frontier innovation.

Let’s create the future of AI.

Experience: 2-5 year of experience.

Key responsibilities

Design and implement Retrieval‑Augmented Generation (RAG) systems

which are much more than chatbots and AI pipelines using vector databases and LLMs. Architect and build MCP pipelines

that connect, coordinate, and supercharge our agentic AI systems. Build and optimize AI/ML models for production, including training, fine‑tuning, quantization, and performance tuning. Deploy and maintain production‑ready AI applications using

MLOps best practices, CI/CD, Docker, and Kubernetes. Monitor your models in the wild, break them down, tune them up, and make them smarter every sprint. Partner with product and engineering teams to quickly build, test, and ship cutting-edge AI features in true startup speed.

Required skills and experience

Strong Python skills and experience with ML/DL frameworks (TensorFlow, PyTorch, Scikit‑learn, Hugging Face Transformers). Have designed, built, and integrated MCP. Practical experience with RAG systems, vector databases, or autonomous agents. Databases: PostgreSQL, MySQL, NoSQL, and vector DBs such as Pinecone, FAISS, Weaviate, or Qdrant. Experience with cloud platforms like AWS (S3, Lambda, RDS, SageMaker), Azure, or GCP. Strong understanding of LLM architectures (Transformers, Encoders/Decoders, Auto‑Encoders). Experience deploying AI systems with MLOps, CI/CD pipelines, Docker, and Kubernetes. Deep proficiency in AI model engineering: training, fine‑tuning, optimizing, and quantizing LLMs and other deep learning models. Familiarity with AI/agentic frameworks such as

LangChain, LlamaIndex, Haystack, LangGraph, AutoGen, CrewAI, Agno,

or similar. Knowledge of custom fine‑tuning techniques:

RLHF, LoRA, QLoRA, PEFT, etc. Comfortable using AI developer tools like Cursor, Codex, Copilot, Gemini in real workflows.

Work Conditions: This is a full-time, (Mon–Sat, 9 AM–6 PM) designed for people who love building in a high-energy, tight-knit startup setting. During the probation period, candidates are required to use their own laptop; an office laptop will not be provided initially.

Culture and values PurpleMerit is an equal-opportunity employer where difference isn’t just accepted—it’s a feature, not a bug. We bring together engineers, dreamers, tinkerers, and rebels to build AI that actually moves the world forward. Inclusivity isn’t a policy for us; it’s our core architecture