Backend Architect - Ontology + Python + AI/ML - Afternoon

Location: Islamabad, lahore, Karachi, Islamabad, lahore, Karachi, Pakistan

Company Overview 

10Pearls is a global, purpose-driven AI-Native digital engineering partner helping businesses re-imagine, ‎digitalize, and accelerate. As an end-to-end digital technology partner, 10Pearls helps businesses create future-proof, ‎transformative ‎digital products that leverage emerging technologies. ‎10Pearls' clients ‎include Global 2000 enterprises, high growth mid-size ‎businesses, and some of the most exciting ‎start-ups from industries like healthcare, fintech, ‎energy, education, ‎real estate, retail, and hi-tech. ‎Headquartered in the Washington DC metro area, 10Pearls has product engineering and ‎software development centers in North America, Latin America, Europe, and South Asia. To learn more, visit https://10pearls.com. 

Role

We are seeking a Backend Architect to lead the design and evolution of a next-generation AI-native platform (AIX), with a primary focus on the Ontology Layer (L3)—the semantic core of the system. This role requires strong architectural ownership and hands-on expertise in building scalable, secure, and multi-tenant backend systems. The ideal candidate will define data models, API contracts, and cross-layer communication patterns while collaborating closely with product, architecture, and engineering teams in an onsite environment in Islamabad. 

Responsibilities:

• Ontology backend architecture — design the multi-tenant PostgreSQL schema (object_types, link_types, object_properties, tool_catalog, action_proposals, audit_events) including Row-Level Security and indexing strategy
• GraphQL API surface — architect the Strawberry-GraphQL API for reads, mutations and subscriptions, and define resolver patterns for tenant isolation and authorisation
• Action Registry & governance — own the propose → simulate → approve → execute state machine, blast-radius classification, dual-approval workflow, and SHA-256 binding hash for approval proofs
• Cross-layer contracts — define the contracts between L2 (data plane), L3 (ontology), L4 (agent runtime) and L5 (governance) — Kafka topic schemas, MCP Truth API, event formats
• Code ownership — write the reference implementation for at least one core module per quarter; review backend PRs; set coding, testing and observability standards
• Mentorship — technically lead a squad of 3–5 backend engineers, pair-program with juniors, run design reviews and architecture decision record (ADR) sessions
• Performance & reliability — set and defend SLOs for L3 services, drive load-testing and capacity planning

Requirements:

• 5–8 years building production backend systems in Python (FastAPI, SQLAlchemy, AsyncIO) or equivalent
• Deep hands-on PostgreSQL: schema design, query optimisation, RLS, pgvector, migrations
• Strong API design skills — GraphQL (Strawberry, Apollo, Hasura) or REST at scale
• Experience with event-driven architectures using Kafka or equivalent (topic design, exactly-once semantics, consumer groups)
• Proven experience owning architecture for a multi-tenant SaaS or platform product
• Strong grasp of distributed-system fundamentals: consistency, idempotency, observability, failure modes
• Kubernetes-native development: containers, health probes, config-as-code, secret management
• Excellent written communication — ADRs, design docs, API specs

Nice to have:

• Prior work on policy-as-code (OPA / Rego) or ABAC authorisation frameworks
• Experience with audit-chain / tamper-evident systems or regulated-industry compliance
• Exposure to LLM-powered agent platforms or tool-use systems
• Experience designing for air-gapped or on-premises deployment
• Contribution to an open-source backend framework or data-platform project