Software Architect Engineer - Islamabad

Job summary

We are seeking a hands-on Software Architect / Tech Lead who can turn business goals into a simple, evolvable architecture and lead engineers to deliver it. You will propose architecture and solution designs, define API contracts and integration patterns, guide developers through implementation, and introduce pragmatic AI/ML capabilities where they create real value.

Responsibilities

• Own the architecture: Service decomposition, domain boundaries, data models, integration patterns (REST/GraphQL/events), and platform guardrails (security, reliability, cost).
• Propose solutions & ship them: Convert product requirements into HLDs, ADRs, sequence diagrams, and implementation plans; align devs and get to production.
• API & integrations: Establish API guidelines (versioning, authN/Z, idempotency, pagination, error contracts); drive contract/consumer testing and backward compatibility.
• Technical leadership: Pair with engineers, review PRs, unblock issues, refine epics/stories, and enforce the definition of done (tests, docs, telemetry).
• AI/ML enablement: Identify high-ROI use cases (search, summarization, classification, recommendations); design RAG pipelines, select providers (OpenAI/Azure/Vertex/Bedrock) or self-hosted inference; define evaluation, guardrails, and observability for AI features.
• Security & privacy by design: Threat modeling, secrets handling, dependency scanning, least-privilege IAM, PII protection, secure SDLC.
• Operational excellence: Define SLOs/SLIs, error budgets, logging/metrics/tracing, release & rollback playbooks with DevOps; performance and cost budgets.
• Roadmap & spikes: Maintain the technical roadmap and tech-debt register; run POCs/spikes to de-risk decisions and keep the stack pragmatic.
• Mentor & communicate: Coach engineers, write clear docs, and collaborate closely with Product and QA on scope and trade-offs.
• AI/ML & Data
• Practical experience integrating LLM features (e.g., RAG, embeddings/vector stores, prompt safety, latency/cost controls) via APIs or self-hosted inference.
• Understanding of MLOps-lite: data pipelines, evaluation/monitoring, drift detection, human feedback loops; partnering with specialists when deep training is required.
• Data architecture literacy: OLTP vs OLAP, CDC/streams, warehousing/lakehouse basics, and governance for PII.

Job description