AI Engineer I
CodeRound is hiring for this role for a VC-backed startup.
- AI Engineer
- Software Engineer
- Agentic AI
- LLMs
- Multi-Agent Orchestration
- RAG
- Tool-Calling
- Prompt Design
- Python
- Evals
- Test Strategies
- Azure
- AWS
- Databricks
- SAP
- Oracle
- Snowflake
- Banking
- MedTech
- CPG
- Industrial
- Client-Facing
- Consulting
- Agentic Tooling
- Open Source
- Eval Frameworks
AI Engineer responsible for designing, building, evaluating, and deploying production-grade agentic AI systems, owning complete workstreams, working with ambiguous requirements, and driving reliable outcomes.
What you'll do
- Design Agentic Workflows: Own the architecture of your workstream — agent topology, tool design, orchestration, and control flow.
- Build Production-Grade Systems: Ship reliable, well-tested agentic code that runs in production environments.
- Design & Own Evals: Build eval harnesses and verification rubrics that hold your workstream to a 99.99% reliability bar.
- Run Client Working Sessions: Lead technical discussions with client engineers and product owners.
- Raise the Team's Bar: Review code from L1 engineers and help them grow into workstream ownership.
Must have
2–4 years of software engineering experience, including at least one system you took to production and owned. Demonstrated experience building with LLMs — multi-agent orchestration, RAG, tool-calling, and prompt design in real applications. Strong Python and solid engineering practice: testing, code review, deployment, and monitoring. Experience designing evals or test strategies for non-deterministic systems, or a clear grasp of why that matters. Ability to work from ambiguous requirements and drive a workstream to a reliable outcome with minimal oversight. Direct, credible communication — you can run a technical conversation with client engineers on your own. Production experience on Azure, AWS, or Databricks. Experience in a regulated enterprise domain — banking, MedTech, CPG, or industrial.
Good to have
Exposure to enterprise systems of record such as SAP, Oracle, or Snowflake. Prior client-facing or consulting-adjacent delivery experience. Contributions to agentic tooling, open source, or eval frameworks.