Staff Engineer, Data Architecture & AI Platform

About You:

You're a real software engineer first. You spent years building and scaling production systems before LLMs made it easy, and a big part of that time was on database, data platform, or infra problems, not a one-year detour. You've since gone deep on AI: agents, tool use, structured outputs, retrieval, memory. You know where a schema, a rule, or a small model beats a prompt, and you don't ship coding-agent output you can't explain.

You're opinionated about structure. When the team wants to ship a new definition of a field that already exists, you're the one who says "no, we use the canonical one, here's why." You're comfortable being the person who slows things down for the right reasons, and you explain your reasoning well enough that people start thinking that way themselves.

You want to stay deep in the technical work rather than move into pure management. You're happy juggling several products at once without a roadmap handed to you, and you give and take blunt technical feedback without it becoming a thing.

You Will:

• Own the data architecture across all GTM apps. That means schema design, object modelling, access patterns, and the path from raw source to what an agent or user reads.
• Build and enforce a data dictionary. Every entity and metric (account health, sentiment score, risk, lifecycle stage) gets one canonical, versioned, deterministic definition that every app reuses. No duplicate definitions and no metrics that are just an LLM's opinion.
• Design our ledger system for scale. Every app keeps metadata ledgers recording the history behind its data: who did what, where a lead came from, what an agent changed. Agents read these ledgers. You'll decide what lives in hot storage and what goes to low-cost cold storage, and how it stays queryable for debugging and audit.
• Bridge R&D and Turbo. Take systems built fast in R&D and re-architect them to run inside Turbo, reconfiguring whatever doesn't fit Turbo's constraints. You'll be the quality gate for what goes to production.
• Design for scale before it hurts. That includes caching layers, indexing and fast read paths, streaming and queueing instead of hammering the DB, and event-driven agent runs instead of nightly crons. You'll keep an active eye on latency, query cost, and token spend.
• Give agents proper data access. Build SQL and query tools, MCP servers, and endpoints so agents fetch exactly what they need instead of dumping data into context. That's how we cut both cost and hallucination.
• Own access and security. Set up RBAC, per-team and per-app views, tenant isolation, PII handling, and secrets management, so customer data only goes where it should.
• Build and harden agents that execute real workflows, with guardrails: prompt injection defense at the tool boundary, human-in-the-loop on anything customer-facing, validated and cited output, and evals that catch regressions before customers do.
• Make the deterministic-vs-probabilistic call on every system, and ship whichever side wins.
• Set the engineering bar. Architecture docs come before implementation. You'll run reviews that teach, write down reusable patterns, and keep runbooks so the same problem never gets solved twice.
• Mentor a fast, young, AI-native team. They ship quickly and lean hard on coding agents. Your job is to make sure they understand the systems they're generating and start thinking in objects, schemas, and contracts.
• Work across multiple products at once (CS, sales, marketing, metrics) and keep the architecture coherent across all of them.

You Have:

• 7+ years of software engineering, mostly pre-LLM. You've built systems from the ground up, not just wired existing ones together.
• Deep data architecture experience. You've modelled structured and unstructured data, designed warehouses and data lakes, built ETL/ELT pipelines, and designed schemas that held up as the product grew.
• Real distributed backend experience. Services that carried significant load, with caching, streaming (Kafka, Pub/Sub or similar), and queueing, and the lessons that come from running them.
• Hands-on cloud depth, ideally AWS. Not just using it, but knowing which services to pick for which data (Athena, S3 tiers, RDS/Postgres and so on) and how to tune them for speed and cost.
• Strong Postgres skills, including performance tuning, indexing, and query design. Supabase is a plus.
• Hands-on production experience with modern LLMs (Anthropic, OpenAI, open models): agents, copilots, or workflows you shipped and maintained.
• Reliability instincts: structured prompting, output validation, evals, observability.
• A security posture that isn't an afterthought: authz, RBAC, secrets, PII, tenant isolation, and awareness of SOC 2 expectations for a B2B SaaS platform.
• Experience leading or mentoring engineers, with a track record of raising the floor of a team.
• Nice to have: Time on a database, data platform, or infra team at scale; having designed a monorepo or internal platform other teams build on; chat-first assistants that take real actions; small language models or decision models replacing LLM calls on high-volume tasks; agent observability and evals in production; vector search and retrieval at scale; CRM and revenue ops data models; real estate or proptech context.
  
  

Our Benefits:

• Competitive Compensation: Attractive salary and benefits package.
• Flexible Work Environment: Fully remote work with flexible hours to promote work-life balance.
• Professional Growth: Opportunities for career advancement and professional development.
• Health & Wellness: Comprehensive health and vision insurance plans.
• Paid Time Off: Generous PTO and paid holidays to recharge and relax.
• Collaborative Culture: A supportive team environment that values innovation and collaboration.
• Equity Options: Opportunity to own a part of Fello and share in our success.
• Cutting-Edge Projects: Work on innovative products that leverage AI and advanced technologies.