Agentic Systems Engineer

Accountabilities:

• • Build modular, plug-and-play AI agents that integrate cleanly into a broader agentic application architecture.
  • Design and implement memory capabilities, including short-term and long-term memory, summarization, and retrieval-backed context.
  • Develop and integrate tools, MCP servers, and reusable agent skills.
  • Build reliable production software with strong separation of concerns, maintainable architecture, and meaningful automated tests.
  • Own the quality of delivered features through testing, evaluations, monitoring, and observability.
  • Analyze production traces, logs, and evaluation results to understand actual system behavior and identify opportunities for improvement.
  • Investigate issues systematically, identify root causes, and implement data-driven fixes rather than relying on assumptions.
  • Contribute to retrieval systems, agent orchestration, context management, and other components of the agentic technology stack.
  • Stay current with emerging agent frameworks, LLM capabilities, and best practices, helping inform technical decisions and future system design.

  Requirements

  • 2–4 years of professional experience developing and maintaining production software.
  • Strong Python skills, including clean, idiomatic syntax, well-structured modules, object-oriented design, separation of concerns, and effective testing practices.
  • Solid understanding of core agentic AI and LLM concepts, including RAG, prompting patterns, tool use, structured outputs, streaming, context management, and generative AI fundamentals.
  • Hands-on experience building a non-trivial application or project using modern agentic or LLM tooling.
  • Ability to quickly understand and navigate unfamiliar codebases.
  • Strong analytical mindset and attention to detail, with a disciplined approach to understanding problems before implementing solutions.
  • Data-driven approach to engineering, with the ability to use production traces, evaluation metrics, and logs to identify meaningful signals and guide decisions.
  • Hands-on experience with Langfuse, LangSmith, or an equivalent LLM tracing and observability platform.
  • Genuine curiosity about emerging AI technologies, agent architectures, frameworks, and industry developments.
  • Experience with search and retrieval technologies such as embeddings, vector databases, hybrid retrieval, and reranking is highly valued.
  • Experience designing and implementing end-to-end LLM evaluations, including evaluation criteria, measurement frameworks, test harnesses, and maintaining reliable scores as models and prompts evolve, is a strong advantage.
  • Deeper experience with LangGraph, including custom graphs, checkpointers, context-management nodes, summarization, windowing, and state pruning, is particularly valuable.
  • Strong communication skills and the ability to explain technical decisions, projects, and trade-offs clearly.

  Benefits

  • Remote work opportunity based in India.
  • Opportunity to work on cutting-edge agentic AI and LLM applications.
  • Hands-on exposure to multi-agent orchestration, MCP, memory systems, retrieval, evaluations, and observability.
  • High level of ownership and trust from the beginning.
  • Fast-moving engineering environment focused on experimentation, measurable outcomes, and continuous learning.
  • Opportunity to work on technically challenging problems at the frontier of generative AI.
  • Structured and rigorous interview process designed to evaluate technical ability, problem-solving, and cultural alignment.
  • Potential for significant professional growth as agentic AI technologies and applications evolve.