Website:
Job details:
This is a high-ownership, product-focused engineering role. You will work at the frontier of agentic AI and LLM workflow automation, designing and building intelligent autonomous agents that power complex business process automation. The ideal candidate combines strong Python engineering skills with hands-on LLM framework experience and a product mindset to deliver reliable, scalable agentic systems.
Key Responsibilities
1. Agentic Workflow Design & Development
Design and develop LLM-based agentic workflows for end-to-end automation of complex business processes using modern orchestration frameworks.
Build multi-step, multi-turn agent pipelines using Lang Chain and Lang Graph, ensuring reliability, observability, and graceful error handling.
Implement tool-use and function-calling patterns enabling agents to interact autonomously with APIs, databases, and external services.
Design agent decision-making logic including planning, reflection, and self-correction loops for long-running workflow automation.
2. RAG Pipelines & Knowledge Systems
Implement and manage Agentic RAG pipelines integrating retrieval systems with agent-based decision making for accurate, grounded responses.
Design and maintain vector database integrations (FAISS, Pinecone, Chroma) for semantic search, document retrieval, and knowledge management.
Curate, chunk, and embed knowledge corpora to support high-quality retrieval for production-grade agent applications.
3. Prompt Engineering & Context Management
Develop robust prompt engineering strategies for accuracy, consistency, and instruction-following across diverse use cases.
Handle context management for long-running and multi-turn agent conversations, including memory modules, context compression, and state persistence.
Design and iterate on system prompts, few-shot examples, and output-structuring techniques to optimize agent behavior.
4. Backend Integration & Platform Engineering
Integrate LLM agent solutions with backend services, REST APIs, relational databases, and data sources to support end-to-end automation workflows.
Collaborate with product and engineering teams to design scalable, maintainable AI feature architectures on the platform.
Support deployment of agent services on cloud environments (AWS / GCP / Azure), including containerization and API serving.
5. Evaluation, Monitoring & Quality Assurance
Implement evaluation frameworks to measure agent accuracy, task completion rate, hallucination rate, and reliability across automated workflows.
Monitor production agent performance, identify failure modes and regressions, and continuously improve agent behavior.
Maintain awareness of LLM model limitations, capability trade-offs, and safety considerations relevant to production deployment.
Click on Apply to know more.