Job description

About Meeru AI Meeru AI is building an AI-native platform that transforms how finance and accounting teams operate. We connect to enterprise financial systems — ERPs, CRMs, billing platforms, HRIS — and apply machine learning to turn fragmented operational data into grounded, auditable intelligence for CFOs, controllers, and FP&A leaders. We deploy on customer terms — SaaS multi-tenant, SaaS single-tenant, and on-premises — across AWS, Azure, and GCP . Our customers are Fortune 500 finance teams who require data isolation, auditability, and compliance. The Role We are looking for a Senior AI Engineer to build the AI and intelligence layer — and help uphold the discipline that keeps it honest. Our output sits adjacent to externally reported financials, so "sounds plausible" is not good enough: everything the AI produces must be grounded in, and traceable to, verified data. You'll build the machine-learning models that learn each customer's patterns, the LLM and agentic systems that produce grounded natural-language output, and help uphold the rigor that keeps that output faithful. You own significant pieces of the layer end-to-end, working closely with our Staff AI Engineer and evaluation engineer, and you build per-customer models without leaking the very signal they're meant to detect. This is a hands-on engineering role. You turn designs into robust production systems, measure quality rigorously, help turn user feedback into durable improvements, and grow toward staff-level technical ownership. • Build per-customer ML models that learn normal vs. anomalous behavior on time-series and operational data, using statistical baselines and gradient-boosted models (LightGBM/XGBoost) with strict anti-leakage discipline. • Build a template-first natural-language generation layer that is strictly grounded in verified data, using guardrails and structured output to prevent hallucinated or unsupported statements. • Implement agentic orchestration (LangGraph or equivalent) with verification and human-confirmation checkpoints. • Integrate managed LLMs running inside customer clouds (AWS Bedrock, GCP Vertex AI, Azure OpenAI). • Partner with an evaluation engineer to gate grounding, hallucination, and faithfulness checks in CI. • Turn user feedback into durable model and system improvements. • Ship production Python services with SQL/warehouse access (Snowflake, PostgreSQL) in containerized, customer-hosted deployments. Nice-to-Have • Finance or accounting domain experience (fintech, FP&A, ERP, audit, financial close). • Feature-attribution experience (e.g. SHAP), along with an understanding of its limits. • Human-in-the-loop systems and feedback loops. • Hands-on work with Bedrock, Vertex AI, or Azure OpenAI in VPC or private deployments. • Zero-inflated or sparse-event modeling.

Requirements

Must-haves

This is a hands-on engineering role. You turn designs into robust production systems, measure quality rigorously, help turn user feedback into durable improvements, and grow toward staff-level technical ownership.
Build per-customer ML models that learn normal vs. anomalous behavior on time-series and operational data, using statistical baselines and gradient-boosted models (LightGBM/XGBoost) with strict anti-leakage discipline.
Build a template-first natural-language generation layer that is strictly grounded in verified data, using guardrails and structured output to prevent hallucinated or unsupported statements.
Implement agentic orchestration (LangGraph or equivalent) with verification and human-confirmation checkpoints.
Integrate managed LLMs running inside customer clouds (AWS Bedrock, GCP Vertex AI, Azure OpenAI).
Partner with an evaluation engineer to gate grounding, hallucination, and faithfulness checks in CI.
Turn user feedback into durable model and system improvements.
Ship production Python services with SQL/warehouse access (Snowflake, PostgreSQL) in containerized, customer-hosted deployments.

Nice-to-haves

Finance or accounting domain experience (fintech, FP&A, ERP, audit, financial close).
Feature-attribution experience (e.g. SHAP), along with an understanding of its limits.
Human-in-the-loop systems and feedback loops.
Hands-on work with Bedrock, Vertex AI, or Azure OpenAI in VPC or private deployments.
Zero-inflated or sparse-event modeling.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aimachine learningawsazuregcpllmmlpythonsql

Platform: workable
Effort: medium