job Responsibilities:
10+ Years
As a Expert Python Developer, you will own the technical implementation of the platform's orchestration layer. You will build the lightweight Python SDK that serves as the primary interface between domain-specific agents (like Lease Abstraction and CAM Reconciliation) and core platform services. Your work will ensure that agents operate within secure, ephemeral sandboxes (Firecracker MicroVMs) while enforcing strict multi-tenant isolation and identity management via WorkOS.
Key Responsibilities
- Platform Control Plane & AI Gateway: Architect and maintain the central platform Control Plane and API Gateway using FastAPI, handling critical functions such as LLM routing, telemetry, and per-tenant cost metering.
- Agentic SDK Engineering: Lead the development of the Tango Python SDK, abstracting complex LLM routing and providing a standardized interface for "Human-in-the-Loop" (HITL) escalations and agent overrides.
- Durable Workflow Orchestration: Implement and manage long-running, multi-stage agentic workflows using Temporal to ensure reliable execution of complex real estate processes.
- Multi-Tenant Backend Services: Build scalable, secure RESTful APIs for the Admin Console, managing complex CRUD operations and ensuring strict logical data isolation across tenants.
- Infrastructure & Security Integration: Collaborate with DevSecOps to deploy backend services via Kubernetes and integrate WorkOS for B2B identity, SSO, and fine-grained Role-Based Access Control (RBAC).
- Data Architecture & MCP: Interface with specialized data stores including PostgreSQL, GraphDB (Neo4j), and Vector Databases (ChromaDB) using the Model Context Protocol (MCP) to ensure agents consume data securely without direct database access.
- LLM Gateway & Cost Controls: Develop and operate the LLM gateway layer (LiteLLM): provider routing and failover (AWS Bedrock, Anthropic), rate limiting, token budget enforcement, runaway-loop kill-switches, and per-run/per-tenant cost metering.
- Production Operations: Participate in SLO definition, incident response, and DR runbook authoring; design for RTO/RPO targets from day one rather than retrofitting.
- Data & Ingestion Pipelines: Build document ingestion pipelines (S3, queues, OCR handoff), schema evolution, and the data contracts through which agents consume tenant data securely , including integration with the client's existing data warehouse and Java/Spring Boot estate.
- Identity & Access Integration: Integrate WorkOS for agent identity, on-behalf-of (OBO) user flows, SSO/SCIM for enterprise tenants, and audit-grade attribution of every agent action and tool call.
Required Skills & Experience
- Backend Mastery: Expert-level proficiency in Python and FastAPI; significant experience building scalable, production-grade RESTful APIs and microservices.
- AI Ecosystem & Agentic Intelligence: Deep experience with LangChain and LangSmith; proven track record building multi-step reasoning chains and tool-use protocols for autonomous agents.
- Cloud & Infrastructure: Strong experience with AWS (Bedrock, ECS, S3) and containerization (Docker/Kubernetes). Understanding of Firecracker MicroVMs for agent sandboxing is a major plus.
- Orchestration & Messaging: Hands-on experience with Temporal for workflow orchestration and RabbitMQ or AWS SQS for event-driven architectures.
- Security & Identity: Proven experience implementing OAuth 2.0 and integrating enterprise identity providers like WorkOS.
- Database Proficiency: Practical experience with relational (PostgreSQL), Graph (Neo4j/GraphDB), and Vector (ChromaDB/PGVector) databases.
- API Design: Demonstrated ability to design and version public-facing APIs and client libraries that other engineering teams depend on.
Desirable
- Prodigy Framework: Familiarity with the Prodigy agentic framework is a significant advantage for this role.
- Domain Expertise: Understanding of Commercial Real Estate (CRE) workflows.
- Evaluation Literacy: Experience building programmatic "Evaluation Harnesses" to measure accuracy and hallucination rates in AI models.
Pay: ₹338,724.32 - ₹4,000,000.00 per year
Work Location: Remote
