Posted today · be early
Senior Data Engineer
FinStrat Management
India, PhilippinesremotePosted 1 day ago
Skill Required
Data-EngineerData-EngineeringCloud-EngineerAI-EngineeringClient-Facing-EngineerSenior-Data-EngineeringSenior-Data-Engineer-JobsSenior-Data-Engineer-PositionsData Engineerdata engineeringGitHub ActionsCloud SecurityAWSEngineeringServerlessautomationsimilar)debuggingsecuritybuildingHubSpotwrittenPythondesignGitISO 27001CI/CDCloudDesign PatternsRustAPIsdbtSQLIAMFulltime
Key highlights
- Compensation commensurate with experience.
- 5+ years of AWS production systems experience, including serverless and CloudFormation.
- Unlimited vacation.
- Hands‑on experience with AI coding tools such as Claude Code.
- Experience integrating accounting/ERP/CRM platforms (QuickBooks, HubSpot, etc.) via APIs.
- Primary technical liaison role, leading client discovery sessions.
Role overview
FSM is building the next generation of AI‑powered financial systems — where accounting, operations, and data applications work together intelligently. As a Senior Data Engineer, you’ll be both a core builder of that platform and a primary technical point of contact for clients as they navigate that transformation.
Responsibilities
- Design, build, and operate FSM's AWS infrastructure: Lambda, API Gateway, SQS, RDS, Cognito, Secrets Manager, CloudWatch.
- Own infrastructure-as-code (CloudFormation), CI/CD (GitHub Actions), and our multi‑tenant security model — IAM roles, tenant isolation, secrets management, auditability.
- Maintain FSM's security and compliance posture — least‑privilege IAM, secrets management, and ongoing SOC 2 / Vanta controls, checks, and audit readiness.
- Maintain and extend our templating/scaffolding system (copier‑based) so new client pipelines and agents deploy fast and stay consistent.
- Build and maintain data pipelines and the S3‑based data platform feeding our agents and BI (DBT, AWS QuickSight).
- Architect and build integrations across accounting, ERP, CRM, and operational apps — QuickBooks, HubSpot, Stripe, Plaid, Box, Slack — into a clean, source‑agnostic layer.
- Ensure consistent, accurate, reconciled data flow.
- Design and code intelligent agents that act on financial data through APIs and natural‑language interfaces, plugging into FSM's orchestration layer.
- Develop the logic frameworks, contracts, and prompts that govern how agents interpret data, take action, and communicate results — including human‑in‑the‑loop controls.
- Test and refine agent performance with real client data — ensuring security, auditability, and alignment with accounting principles.
- Client‑facing: Walk clients through agent behavior and outputs directly, building their confidence and trust in the automation.
- Client‑facing: Lead discovery sessions with client accounting, finance, and operations teams to understand their workflows, data challenges, and automation needs.
- Serve as the primary technical liaison between FSM and the client — explaining findings, options, and trade‑offs in terms non‑technical stakeholders can act on.
- Identify opportunities where AI agents can replace repetitive processes, improve reconciliation, or enhance insight delivery.
- Translate accounting concepts into technical requirements — and technical constraints back into business terms.
- Create comprehensive documentation of automations, agent behaviors, infrastructure, and client‑specific implementations.
- Establish repeatable playbooks, naming conventions, and frameworks that take us from trial‑and‑error to scalable structure.
- Partner with leadership to define best practices as the department evolves.
- Share progress, blockers, and technical decisions clearly — and early — across Engineering, Client Delivery, and CSM, not just with the AI Agent Engineering team.
Requirements
- 5+ years building and operating production systems on AWS, with real depth in serverless (Lambda/API Gateway), infrastructure‑as‑code (CloudFormation or similar), and cloud security (IAM, secrets, multi‑tenancy).
- Strong data engineering: Python, SQL, data pipelines, and a modern data stack (DBT, S3/data‑lake patterns, a BI tool).
- Solid CI/CD and Git discipline; you write reviewable code and review others' well.
- Experience integrating accounting/ERP/CRM platforms (QuickBooks, HubSpot, or similar) via APIs, webhooks, or connectors.
- Comfort with LLM‑based tools and APIs — prompt design and evaluating agent outputs for accuracy and reliability.
- Hands‑on experience using AI coding tools (e.g., Claude Code) as part of your daily development workflow — planning, building, reviewing, and documenting code with them.
- Demonstrated experience working directly with clients or business stakeholders — leading discovery, presenting findings, managing expectations, not just coding behind the scenes.
- You take positions and own decisions. Under ambiguity you form a recommendation, explain your reasoning, and drive it — rather than waiting to be told.
- You communicate clearly under pressure — status, trade‑offs, and blockers surfaced early — with technical and non‑technical audiences alike.
- Excellent verbal and written English; strong overlap with US Eastern hours.
Nice to have
- Experience building LLM/AI agents or orchestration systems.
- Templating tools (copier/cookiecutter).
- Working knowledge of accounting (reconciliations, journal entries, chart of accounts, financial reporting) — or a genuine willingness to learn the domain; we'll teach it.
- Experience taking a team from bespoke work toward standardized, documented process.
- Familiarity with SOC 2 / Vanta‑style compliance.
Benefits
- Compensation commensurate with experience
- Unlimited vacation
- Ongoing education and training
- Bonuses
Additional details
- Our Stack: AWS (Lambda, API Gateway, SQS, RDS + RDS Proxy, Cognito, Secrets Manager, CloudFormation, CloudWatch) · Python · GitHub Actions · Claude Code · copier · DBT · S3 · AWS QuickSight · LLM agents (Claude) · QuickBooks, HubSpot, Stripe, Plaid, Box, Slack
- What Success Looks Like:
- Clients trust you as a technical partner, not just a vendor resource - they bring you problems before they bring you requirements.
- Discovery sessions consistently surface automation opportunities that convert into shipped agents and integrations.
- The platform you build is secure, multi‑tenant, and reliable - deployments are repeatable and don't bottleneck on any one person.
- Internal teams (Engineering, Client Delivery, CSM, Leadership) have clear visibility into your work through documentation and proactive communication.
- Agent deployments are accurate, auditable, and aligned with sound accounting principles — earning continued client confidence.
- Security and compliance stay audit‑ready - SOC 2 / Vanta controls and checks are maintained without last‑minute fire drills.
- Originally posted on Himalayas