Andiamo
Agentic Systems Engineer - Interactive AI Deep-Research Platform
Andiamo, New York, New York, us, 10261
Agentic Systems Engineer - Interactive AI Deep-Research Platform
Join a fast-moving team building the backend intelligence that powers autonomous AI research agents for the financial sector. You will architect the orchestration layer that coordinates multiple specialized agents, manages complex multi-step workflows, and executes mission-critical analysis with reliability, observability, and security at enterprise scale. About The Role Why This Role Matters: Real impact: Your systems will drive autonomous research used by leading financial institutions. Deep engineering: Tackle hard problems in distributed systems, agent coordination, and real-time data processing. AI in production: Ship multi-model, cost-aware AI workflows that combine language models with streaming market data. Your First 90 Days: Design and launch a production-grade multi-agent workflow for real financial research tasks. Build an agent orchestration service (state machines, task queues, retries, SLAs) with senior mentorship. Own a core subsystem end-to-end—from architecture and implementation through deployment and on-call. Demonstrate measurable reliability and latency improvements across an autonomous research pipeline. What You’ll Build: Agent Orchestration & Workflow Engines: Coordination services that route work across specialized agents; stateful execution with backoff, deadlines, and idempotency. Multi-Agent System Architecture: Communication patterns, task delegation, result synthesis, and dynamic resource allocation for heterogeneous workloads. Autonomous Task Execution: Long-running, multi-phase flows with structured observability, failure isolation, and graceful degradation—while preserving human oversight. AI Model Integration: Pluggable interfaces for LLMs and domain models (routing, fallback, A/B testing, cost/latency budgets) with easy swap-in/swap-out. Real-Time Data Pipelines: Event-driven ingestion of streaming market data, regulatory filings, news, and alternative datasets that trigger agent workflows. Agent Memory & Context: Vector search and knowledge graphs to persist context, reuse prior analysis, and improve retrieval over long sessions. Enterprise-Grade Reliability: Authentication, audit logging, policy controls, and compliance-ready telemetry for regulated customers. Must-Have Qualifications 5+ years building production-scale distributed systems and backend services. Strong CS fundamentals: algorithms, concurrency, networking, storage, and system design. Experience with agent frameworks or multi-agent design patterns (coordination, specialization, choreography). Expertise in Python, Node.js, or Rust; hands-on with microservices and event-driven architectures. Practical experience with vector databases and semantic retrieval (e.g., Pinecone, Weaviate, Chroma) and embedding-based search. Track record of shipping systems that handle mission-critical workloads and real user traffic. Clear technical communication and the ability to collaborate across product, data, and enterprise stakeholders. Nice to Have Background building AI/ML production systems, workflow orchestration, or autonomous agent platforms. Familiarity with financial data and APIs or enterprise integration patterns. Experience with Kafka (or similar), Kubernetes, and infrastructure-as-code. Technical leadership: mentoring, architectural decision-making, and setting engineering standards. Startup experience or substantial side projects built from scratch. Tech You Might Use Languages & Runtime: Python, Node.js/TypeScript, Rust Workflow & Messaging: Kafka, task queues, state machines AI Stack: LLM providers, routing/fallback, evaluation/A&B, prompt tooling Retrieval: Vector DBs, knowledge graphs, embedding pipelines Data: Streaming ingestion for market/news/filings; event-driven architectures Platform: Kubernetes, containers, observability (metrics, logs, traces), IaC Security: AuthN/AuthZ, audit, policy enforcement for enterprise environments How We Work Ownership: Engineers own outcomes end-to-end—design, ship, measure, iterate. Evidence-driven: We prioritize telemetry, experimentation, and operational feedback loops. Safety & Reliability: Guardrails and fallbacks are first-class citizens, not afterthoughts. What’s In It for You Build foundational agentic infrastructure at the frontier of AI and finance. Ship systems that augment human decision-making at enterprise scale. Work with a pragmatic team that values clarity, velocity, and craftsmanship. Andiamo is an equal opportunities employer and welcomes applications from diverse candidates.
#J-18808-Ljbffr
Join a fast-moving team building the backend intelligence that powers autonomous AI research agents for the financial sector. You will architect the orchestration layer that coordinates multiple specialized agents, manages complex multi-step workflows, and executes mission-critical analysis with reliability, observability, and security at enterprise scale. About The Role Why This Role Matters: Real impact: Your systems will drive autonomous research used by leading financial institutions. Deep engineering: Tackle hard problems in distributed systems, agent coordination, and real-time data processing. AI in production: Ship multi-model, cost-aware AI workflows that combine language models with streaming market data. Your First 90 Days: Design and launch a production-grade multi-agent workflow for real financial research tasks. Build an agent orchestration service (state machines, task queues, retries, SLAs) with senior mentorship. Own a core subsystem end-to-end—from architecture and implementation through deployment and on-call. Demonstrate measurable reliability and latency improvements across an autonomous research pipeline. What You’ll Build: Agent Orchestration & Workflow Engines: Coordination services that route work across specialized agents; stateful execution with backoff, deadlines, and idempotency. Multi-Agent System Architecture: Communication patterns, task delegation, result synthesis, and dynamic resource allocation for heterogeneous workloads. Autonomous Task Execution: Long-running, multi-phase flows with structured observability, failure isolation, and graceful degradation—while preserving human oversight. AI Model Integration: Pluggable interfaces for LLMs and domain models (routing, fallback, A/B testing, cost/latency budgets) with easy swap-in/swap-out. Real-Time Data Pipelines: Event-driven ingestion of streaming market data, regulatory filings, news, and alternative datasets that trigger agent workflows. Agent Memory & Context: Vector search and knowledge graphs to persist context, reuse prior analysis, and improve retrieval over long sessions. Enterprise-Grade Reliability: Authentication, audit logging, policy controls, and compliance-ready telemetry for regulated customers. Must-Have Qualifications 5+ years building production-scale distributed systems and backend services. Strong CS fundamentals: algorithms, concurrency, networking, storage, and system design. Experience with agent frameworks or multi-agent design patterns (coordination, specialization, choreography). Expertise in Python, Node.js, or Rust; hands-on with microservices and event-driven architectures. Practical experience with vector databases and semantic retrieval (e.g., Pinecone, Weaviate, Chroma) and embedding-based search. Track record of shipping systems that handle mission-critical workloads and real user traffic. Clear technical communication and the ability to collaborate across product, data, and enterprise stakeholders. Nice to Have Background building AI/ML production systems, workflow orchestration, or autonomous agent platforms. Familiarity with financial data and APIs or enterprise integration patterns. Experience with Kafka (or similar), Kubernetes, and infrastructure-as-code. Technical leadership: mentoring, architectural decision-making, and setting engineering standards. Startup experience or substantial side projects built from scratch. Tech You Might Use Languages & Runtime: Python, Node.js/TypeScript, Rust Workflow & Messaging: Kafka, task queues, state machines AI Stack: LLM providers, routing/fallback, evaluation/A&B, prompt tooling Retrieval: Vector DBs, knowledge graphs, embedding pipelines Data: Streaming ingestion for market/news/filings; event-driven architectures Platform: Kubernetes, containers, observability (metrics, logs, traces), IaC Security: AuthN/AuthZ, audit, policy enforcement for enterprise environments How We Work Ownership: Engineers own outcomes end-to-end—design, ship, measure, iterate. Evidence-driven: We prioritize telemetry, experimentation, and operational feedback loops. Safety & Reliability: Guardrails and fallbacks are first-class citizens, not afterthoughts. What’s In It for You Build foundational agentic infrastructure at the frontier of AI and finance. Ship systems that augment human decision-making at enterprise scale. Work with a pragmatic team that values clarity, velocity, and craftsmanship. Andiamo is an equal opportunities employer and welcomes applications from diverse candidates.
#J-18808-Ljbffr