Qentelli - AI Engineer

GENTIC AI ENGINEER

About the Role :

Qentelli is building the next generation of Agentic AI systems that transform how enterprises develop, test, and operate software. As an Agentic AI Engineer, you will design and build production-grade AI agents using LangGraph, LangChain, and LangSmith. You will work at the intersection of LLM engineering, enterprise integration, and autonomous system design, helping engineering teams ship faster with AI that genuinely understands their workflows.

Key Responsibilities :

Agent Design & Development :

- Design and implement multi-step LangGraph agents.

- Build stateful, graph-based agent workflows with conditional routing, parallel execution, and human-in-the-loop checkpoints

- Integrate LangChain chains, tools, and retrievers into agent pipelines

LLM Engineering & Prompting :

- Design and optimise prompts for structured output, schema-conformant data generation, and reasoning tasks

- Implement context management strategies to control token usage and cost

- Evaluate and select appropriate LLMs (OpenAI, Anthropic, Gemini, open-source) per task requirements

- Build RAG pipelines with vector stores for domain-specific knowledge retrieval

Observability & Quality :

- Instrument all agents with LangSmith tracing, evaluation datasets, and automated regression tests

- Build dashboards and alerts for agent performance, latency, cost, and failure modes

- Implement human-in-the-loop review workflows and approval gates

- Define quality metrics and run structured evals before production deployment

Enterprise Integration :

- Connect agents to enterprise toolchains.

- Build secure, governed agent deployments - API key management, access controls, audit logging

- Collaborate with client engineering teams to scope, build, and iterate on agent solutions

Platform & Reusability :

- Contribute to Qentelli's internal agent library

- Build reusable agent components, tools, and prompt templates across projects

- Document agent architectures, design decisions, and runbooks for client handover

Required Qualifications :

Core Technical Skills :

- 4+ years of Python development; strong understanding of async, concurrency, and API patterns

- 2+ yrs in LangChain and LangGraph

- Demonstrated ability to design graph-based agent workflows with branching, loops, and state management in LangGraph

- Experience with LangSmith for tracing, debugging, and evaluating LLM applications

- Proficiency with LLM APIs : OpenAI, Anthropic Claude, or equivalent

- Strong prompt engineering skills - structured outputs, chain-of-thought, few-shot, and tool use patterns

- Experience integrating agents with external APIs and databases via REST or SDK connections

Systems & Engineering :

- Solid understanding of software engineering practices : Git, CI/CD, testing, code review

- Solid understanding with vector databases (Chroma, Pinecone, pgvector) and embedding models, re-ranking and RAG implementation.

- Experience with containerisation (Docker) and cloud deployments (AWS, Azure, or GCP)

- Ability to read and work with enterprise API documentation and data schemas

Preferred Qualifications :

- Experience building multi-agent systems or agent-to-agent communication patterns

- Knowledge of evaluation frameworks : RAGAS, LangSmith Evals, or custom harnesses

- Contributions to open-source LLM or agent projects