AI Infra Engineer

Serotonin•Toronto, NY

1d•Remote

About The Position

We're building an intelligence layer at Serotonin, and we need someone to own it end to end. This role sits at the intersection of competitive intelligence, data engineering, and agentic AI. You'll pull signals from social platforms, on-chain activity, and paid media, and turn them into structured intelligence that shapes how we run accounts and campaigns, both internally and for clients. This is a systems role, not a workflow automation role. You'll design and maintain the data pipelines, agentic workflows, and full-stack applications that make our intelligence capabilities real and repeatable. The best outcome isn't a one-off solution for one team, rather it looks like a system three teams depend on independently six months from now. You'll work directly with marketing leads and client teams, understand how they actually operate before you build anything, and ship things that are live, used, and measurable.

Requirements

Software Engineering (non-negotiable)
Python, fluently: pipelines, agent systems, data processing, scripting
Active GitHub with meaningful commit history, not forks and tutorials
REST and GraphQL API experience building integrations, not configuring connectors
SQL: schema design, not just pulling rows
Full-stack development: Next.js or equivalent; you've shipped something users interact with in production
LLM API usage: building directly with Anthropic or OpenAI APIs, including agent orchestration frameworks (LangGraph, CrewAI, or custom)
Vector databases and RAG: embeddings, retrieval pipelines, semantic search
Intelligence and Signal Layer Experience ingesting and fusing data from multiple external sources: social APIs (X, LinkedIn), ads APIs (Meta, Google), or similar multi-source environments
Ability to translate raw signal into structured, actionable output for non-technical stakeholders
Experience building dashboards or reporting layers others depend on operationally, not just internal notebooks
AI Tooling No-code tools (Make, n8n, Zapier) are part of your toolkit but they don’t act as the foundation of your practice. You reach for them when they're genuinely the fastest path to a reliable outcome. A strong developer who knows when to use no-code will always outperform a no-code specialist who can't own the harder technical work.

Nice To Haves

On-Chain Data (strong plus)
Dune Analytics or equivalent SQL querying against on-chain data
Familiarity with wallet activity, token flows, and what on-chain behavior signals about an entity or protocol
Not required, but you should be the kind of person who picks up a new data domain fast
Domain Working knowledge of how GTM and marketing teams operate day to day: PR, content, growth, social. Enough to know where intelligence systems create real leverage.
Crypto experience is a meaningful plus.

Responsibilities

Build and maintain agentic systems that continuously collect, process, and surface signals from social platforms, on-chain data, and ads ecosystems.
Turn raw, multi-source data into structured intelligence that informs real account and campaign decisions.
Design and maintain the pipelines that power those systems: ingesting from multiple external APIs, enriching and modeling data, and building semantic layers that both humans and AI agents can actually use.
Build multi-step AI agents that automate research, synthesis, monitoring, and reporting. Not single-call LLM wrappers, but orchestrated systems with memory, defined tools, and structured output schemas that teams can depend on.
Build full-stack applications and dashboards that make intelligence accessible across client accounts. Live tools, not decks.
Spend real time with internal teams across marketing, PR, content, growth, and social before writing a line of code. Map how work actually moves, identify where signal gaps create friction, and build systems that solve for multiple teams from a single architecture.
Stay ahead of the AI tooling landscape. Know when a no-code connector is the right call and when you need to build.
Define how impact gets measured, and build the reporting to prove it.