Staff II Software Engineer

BlackLine Systems, Inc.; Bengaluru; On-site; INR 6,000,000 - 9,000,000; Full time

Job summary

BlackLine Systems, Inc. in Bengaluru seeks a Senior Staff Engineer with 15+ years of software engineering for a hands-on, architectural leadership role in Agentic AI and Generative AI systems.

You will drive design, deliver 3-4 end-to-end AI implementations, and guide Staff Engineers and Tech Leads across the AI platform to production readiness. You will architect scalable AI systems, define governance, and ensure secure, compliant financial workflows, while mentoring the team and coordinating

Qualifications

• 15+ years of software engineering experience
• Hands-on AI systems experience with end-to-end delivery
• 3-4 end-to-end AI implementations from design to production
• Experience in FinTech or enterprise compliance patterns

Responsibilities

• Own 3-4 AI delivery cycles from discovery to production
• Lead architecture reviews, ADRs, and design governance
• Mentor Staff Engineers and Tech Leads to grow in agent governance
• Define engineering standards for agentic AI and multi-agent systems
• Collaborate across Finance, Compliance, and Platform teams
• Present proposals and risk assessments to leadership

Skills

Software engineering

Generative AI

Agentic AI

Python

System architecture

Cloud platforms

Education

MS in Computer Science

Tools

GCP

AWS

Kubernetes

Vertex AI

Temporal

Job description

Make Your Mark:

We are looking for an accomplished, deeply hands‑on Senior Staff Engineer with 15+ years of software engineering experience and a focused track record in Agentic AI,Generative AI systems and Software/Data engineering. This is a senior individual contributor role that combines architectural leadership with direct execution - you will drive the design and delivery while simultaneously guiding and unblocking a team of Staff Engineers and Technical Leads. The ideal candidate has independently driven 3-4 end‑to‑end AI implementations - from requirements gathering and solution design through to production deployment - and is equally comfortable whiteboarding complex multi‑agent architectures and writing production‑grade Python. You bring the experience to set the technical direction, the judgment to make critical architecture decisions, and the hands‑on depth to get into the code when the team needs unblocking. This role demands breadth and depth across the stack: you will architect scalable AI systems, build and orchestrate intelligent agents, embed robust agent governance frameworks, and establish the engineering standards that make those agents enterprise‑ready for a secure and compliant financial platform.

You’ll Get To:

Own 3-4 complete AI delivery cycles: discovery, requirements, solution design, build, test, and production deployment on GCP or any cloud.

Translate ambiguous business problems into concrete solutions with clear success metrics.

Define and manage delivery milestones across cross‑functional stakeholders including Finance, Compliance, and Platform Engineering.

Lead architecture decision records (ADRs), design reviews, code reviews, and production readiness assessments.

Set quality bars for the team: performance benchmarks, eval thresholds, observability standards, and security posture.

This role carries explicit responsibility for the technical growth and day‑to‑day unblocking of Staff Engineers and Technical Leads working across the Agentic AI platform.

Architecture Decomposition:

Break down the platform into clear workstreams and assign ownership to Staff Engineers with explicit API contracts and interface definitions.

Hands‑on Unblocking:

When Staff Engineers or Tech Leads are stuck on system design decisions, integration challenges, or production issues step in directly with code, design sketches, or paired debugging sessions. This is not a review‑only role.

Design Guidance:

Review component‑level designs proposed by Staff Engineers, identify gaps in governance, observability, or compliance thinking, and guide them to better solutions without simply overriding.

Standards Setting:

Define and document engineering standards for the agentic AI domain: agent identity patterns, audit trail requirements, HITL design principles, Kafka topic design, and eval harness setup.

Mentorship:

Run architecture working sessions, design clinics, and 1:1 coaching with Staff Engineers and Technical Leads to build their depth in agent governance and FinTech compliance patterns.

Cross‑Team Influence:

Align architecture decisions across Engineering, security, Compliance, and Data Platform teams. Represent the AI Platform in cross‑functional technical forums.

Architect the Agentic AI System:

agent registry, policy engine, metrics ingestion pipeline, anomaly detection, and HITL orchestration.

Design multi‑agent pipelines with clear trust tier boundaries (T0‑T3), separation of concerns, fault isolation, and full observability.

Define data flows, integration patterns (REST, gRPC, SSE, event‑driven), and storage strategies for AI workloads in a multi‑tenant FinTech environment.

Design the durable DAG execution, pause/resume for long‑running financial close workflows, event‑sourced audit trail, and SLA‑tracked HITL approval gates.

Present technical proposals, architecture trade‑offs, and risk assessments to senior leadership and key stakeholders.

Design, build, and deploy intelligent agents for complex, multi‑step financial reasoning tasks.

Implement agent orchestration using LangGraph , Temporal, Google Workflows, or equivalent.

Deploy agents on GCP (Vertex AI Agent Builder, Cloud Run, GKE) or AWS / Azure equivalents; establish K8s namespace isolation for T1 certified partner agents.

Implement MCP (Model Context Protocol) server and client patterns for external agent integration (T2/T3 trust tiers via SSE/HTTPS).

Implement ReAct, Plan‑and‑Execute, and other advanced prompting and reasoning strategies suited to accounting domain tasks.

Design and implement RAG pipelines with vector stores (Pinecone, Weaviate, pgvector) for domain‑specific financial document retrieval.

Fine‑tune, prompt‑engineer, and evaluate foundation models (Gemini, GPT‑4, Claude) for accounting‑domain tasks including close commentary generation and variance explanation.

Optimise inference costs through caching, batching, model routing, and quantisation strategies; maintain token cost visibility in Control Tower.