Job Description & Details
This is a hands-on AI engineering role focused on moving beyond simple chatbot wrappers and actually building autonomous agentic workflows for SDLC automation. If you enjoy wiring up LLMs with execution harnesses, state machines, and real backend systems, this is a solid opportunity to build production-grade agent loops.
What You'll Actually Be Doing
You will spend your days designing and coding autonomous agent systems that can plan, act, observe, and self-correct across complex software development lifecycles. Expect to build robust backend services that orchestrate multiple LLM calls, manage vector databases, and tie directly into enterprise APIs and toolsets. The real challenge here isn't just writing the prompt, but making the agent loop reliable, scalable, and capable of handling messy real-world edge cases without hallucinating or failing silently.
The Core Tech Stack
You need to be completely fluent in either Python (with FastAPI) or TypeScript, paired with hands-on experience using LLM frameworks like LangChain, LangGraph, or Semantic Kernel. A deep understanding of RAG architectures, vector databases, and prompt engineering is non-negotiable since you'll be building the context pipelines that feed these agents. Experience deploying these workloads on AWS and setting up solid observability and logging will separate you from other applicants.
Interview Expectations
Expect the hiring team to grill you on how you handle state management and recursion depth in complex agentic loops like LangGraph. They will likely ask you to debug a scenario where an agent gets stuck in an infinite planning-acting loop or fails to properly handle API rate limits and tool execution errors. They want to see that you understand the brittle nature of LLMs in production and know how to build deterministic guardrails around stochastic outputs.
Application Advice
To get past the ATS and grab the recruiter's attention, make sure your resume explicitly highlights production-level Python or TypeScript backend development alongside specific LLM orchestration frameworks like LangChain or Semantic Kernel. Don't just list keywords—include metrics or concrete examples of RAG pipelines, vector databases, and cloud deployments you have successfully shipped.