Senior Applied AI Engineer

This is a fully remote position, open to applicants in United States, +1 more state.

📋 Description

• Develop and launch Tango's inaugural AI-driven product.

• Design, create, and deploy production AI agents utilizing LangGraph.

• Take full ownership of agents from start to finish, encompassing graph design, tool specifications, prompt and context engineering, reliable execution, failure and retry mechanisms, as well as cost and latency considerations.

• Construct agent tools interfacing with internal systems over MCP, employing direct API interactions and agent-to-agent communications where suitable.

• Implement human-in-the-loop review processes featuring interruption points, confidence indicators, and correction pathways.

• Develop and refine retrieval strategies, including chunking, hybrid retrieval, grounding, and citation to the source page and paragraph.

• Contribute to agent assessments utilizing golden datasets, LLM-as-judge methodology, deterministic scoring systems, and CI regression testing suites.

• Identify quality failures and rectify issues related to retrieval, prompts, tooling, models, or ground-truth data.

• Manage safety protocols at the agent level, including resistance to prompt injection, handling of PII, refusal mechanisms, and escalation processes.

• Collaborate with Product teams to translate accuracy benchmarks, confidence disclosures, and human-in-the-loop triggers into deliverable behaviors.

• Coordinate with Platform Engineering for deployment and with the Data Platform for curated datasets.

• Integrate agent session and utilization analytics into the data warehouse.

• Transition reference agents to domain teams and contribute to the establishment of a shared agent quality standard.

⛳️ Requirements

• Over 7 years of professional experience in software engineering.

• At least 2 years of experience developing LLM-powered systems that have been deployed to production for real users.

• Proficient in Python and its associated service stack, including FastAPI, Pydantic, or similar frameworks.

• Familiar with testing methodologies, code reviews, CI/CD processes, and ownership of production environments.

• Hands-on experience with an agent orchestration framework, with a strong preference for LangGraph.

• Practical knowledge of at least one frontier model API.

• Experience with LLM evaluation processes, including golden datasets, LLM-as-judge, deterministic scoring, regression testing, and release gating.

• Knowledge of MCP tool servers or similar tool and function-calling protocols.

• Familiarity with multi-agent patterns.

• Production experience in RAG and retrieval practices, including chunking, hybrid retrieval, grounding, citation, and troubleshooting retrieval failures.

• Understanding of LLM observability and tracing tools, such as LangSmith, Langfuse, Arize, or equivalents.

• Experience with managing prompts and version control.

• Good judgment regarding hallucinations, prompt injections, and silent performance degradation.

• Preferred: experience with graph-backed agent memory or knowledge graphs like Neo4j.

• Preferred: familiarity with production vector and hybrid retrieval stores, such as pgvector, Pinecone, Weaviate, or Qdrant.

• Preferred: experience in asynchronous task orchestration for long-running document workflows, like Celery/Redis or similar.

• Preferred: knowledge in document intelligence and large-scale information extraction, including OCR, layout-aware parsing, and structured extraction from lengthy documents.

• Candidates must be authorized to work in the U.S. for any employer.

• Tango is currently unable to sponsor employment-based visas.

🏝️ Benefits

• Comprehensive health, dental, and vision insurance.

• 401(k) plan with company matching contributions.

• Generous paid time off policy.

• Fully remote work environment.

• Inclusive and collaborative workplace culture.

• Equal opportunity employment practices.