Posted today · be early
Forward-Deployed AI Engineer
Saaf Finance
IndiaremotePosted 1 day ago
Skill Required
Forward-Deployed-AI-EngineerForward-Deployed-ML-EngineerForward-Deployed-AI-EngineeringForward-Deployed-EngineerForward-Deployed-Software-EngineerSenior-Forward-Deployed-EngineerAI-EngineerAI EngineerAIData StructuresSystem DesignEngineeringautomationObservabilitysimilar)buildingfintechPythondesignCloudjQueryAPIsSassRAGAWSandGenerative AIGoMachine LearningFulltime
Key highlights
- Competitive compensation
- Unlimited PTO
- $2,000/year professional development budget
- Must have shipped AI‑powered products or agents to production
- Strong Python proficiency required
- Experience in fintech or regulated industries preferred
Role overview
Saaf AI, part of American Heritage Lending, is building the future of mortgage lending by combining cutting‑edge AI with proven lending operations. The company is hiring a Forward‑Deployed AI Engineer to design and ship AI‑powered products and agents end‑to‑end, owning the full arc from business problem to a working AI system in production.
Responsibilities
- Design and ship AI‑powered products and agents from end‑to‑end, owning the full arc from business problem to production system.
- Automate underwriting decisions and build intelligent borrower‑facing experiences.
- Create agents that navigate complex loan workflows and compress multi‑day tasks into minutes.
- Build multi‑step LLM agents that can navigate loan files, evaluate conditions, flag exceptions, and surface recommendations to underwriters.
- Design agentic workflows with tool use, structured output, memory, and human‑in‑the‑loop checkpoints.
- Integrate AI agents with internal systems, external data sources, and third‑party APIs to close the loop between insight and action.
- Extend and improve document extraction pipelines for income documents, appraisals, title, and more using the latest extraction and classification models.
- Build retrieval‑augmented systems that let underwriters and borrowers query complex loan files in natural language.
- Design feedback mechanisms that use reviewer corrections to improve extraction accuracy over time.
- Build AI‑generated task lists and smart checklists that adapt to each borrower’s loan scenario.
- Create intelligent routing and prioritization to determine which loans need attention, which borrowers are at risk of drop‑off, and which conditions are blockers.
- Design conversational AI experiences that guide borrowers through complex decisions without requiring a loan officer on every interaction.
- Build evaluation frameworks that measure AI output quality against ground truth, not just latency.
- Design human review workflows where AI confidence is low and use those reviews to create labeled training data.
- Monitor production AI performance—including accuracy, coverage, failure modes—and iterate relentlessly.
- Ship early, measure, and improve with short feedback loops, driving features end‑to‑end (requirements, design, implementation, deployment, monitoring).
Requirements
- Proven experience shipping LLM‑powered features or agents to production, handling real‑world problems such as latency, hallucination, evaluation, prompt regression, and retrieval quality.
- Strong Python skills; ability to write clean, production‑quality Python, comfortable with async, APIs, and data structures.
- Fluency with major LLM APIs and at least one agent orchestration framework; practical knowledge of structured output, tool calling, RAG, and evaluations.
- Product thinking: ability to translate a business process into an AI system design with user‑centric focus.
- Ownership without guardrails: drive features end‑to‑end (requirements, design, implementation, deployment, monitoring) without handing off hard parts.
- Fast iteration mindset: ship early, measure, improve, and thrive on short feedback loops.
Nice to have
- Experience in fintech, lending, insurance, or other regulated industries.
- Full‑stack experience, comfortable building lightweight UIs, APIs, or internal tools.
- Cloud deployment experience (AWS preferred), including containerization, deployment, and operation.
- Experience with document understanding (OCR, extraction models, structuring data from unstructured sources).
- Background in building human‑in‑the‑loop systems (review workflows, feedback capture, data flywheel design).
- Experience with mortgage or real estate data.
- Familiarity with workflow automation tools (e.g., n8n or similar).
- Exposure to fine‑tuning or model adaptation for domain‑specific tasks.
- Experience building AI products that handle PII and comply with regulatory requirements such as FCRA and GLBA.
Benefits
- Competitive compensation
- Unlimited PTO
- Remote‑first with flexible hours
- $2,000/year professional development budget
- Home office setup stipend
Additional details
- Saaf AI is part of American Heritage Lending, a top‑10 private lender processing billions in loan volume with 15+ years of growth.
- Backed by large asset managers and funds; company is growing fast.
- The company is actively deploying AI across underwriting, document processing, and borrower experience.
- Mortgage lending is one of the last industries to be meaningfully transformed by AI; the problems are hard, data is rich, stakes are real.
- AI‑first culture: engineers use AI to build AI (Claude Code, agentic workflows, AI‑assisted code review).
- Direct impact on a $2 trillion industry; work touches every loan processed.
- Why Saaf narrative describing the significance and impact of the role.
- Originally posted on Himalayas.