The firm is looking to build a platform that will enable the next generation of client service, focusing on a distinctive combination of AI capabilities, domain expertise, and human judgment. This involves developing exciting technology for the firm by designing, building, and operating agentic systems that power GenAI applications.
Requirements
- Proven agentic systems experience with LangGraph/LangChain (function/tool calling, state management, planning, retries/fallbacks)
- Orchestrating ETL/DAG workflows (Airflow, Prefect, Dagster, AWS Step Functions) with idempotency, backfills, SLAs, lineage
- Strong Python (asyncio, FastAPI, Pydantic); TypeScript/Node.js a plus
- Expertise in React and Next.js (SSR/ISR, streaming UIs)
- SQL mastery with PostgreSQL/RDS; solid schema design and query optimization
- Retrieval & data layer: vector search (pgvector/FAISS/Pinecone) and knowledge graphs (SPARQL/Gremlin) for RAG/memory
- CI/CD with GitHub Actions; GitHub PR workflows; Docker/Kubernetes; AWS (ECS/Fargate, Lambda, S3, CloudWatch, IAM)
Responsibilities
- Design, build, and operate agentic systems that power GenAI applications.
- Use LangGraph and LangChain to author multi-step, tool-using agents with robust state management, planning, and recovery.
- Define and evolve function/tool schemas.
- Implement tool routing, parallelization, and fallbacks.
- Integrate retrieval, memory, and context windows.
- Add guardrails, evaluation harnesses, and tracing for safety and observability.
- Expose agents via APIs and workflows.
Other
- Thrives in a high performance/high reward culture.
- Resilience to try another way forward.
- Continuous learning and embracing fast-paced learning.
- Owning your journey.
- Strong problem-solving, ownership mindset, and comfort in a fast-paced, collaborative environment.
