Product Manager

Okara is an AI CMO that runs a team of specialized agents across every organic and social channel that matters for growth. We've built an agent for SEO, one for GEO (AI visibility), and an AI writer for organic content, and we're looking for a product manager to help us make these agents, and the results they produce, better.
The Role
You'll sit inside engineering and own both what we build and how it ships. You'll turn a fast-growing list of new agents, customer requests and quality issues into a clear roadmap, then run the cadence that gets it to production every week.
This is a hybrid role. About half is the product: deciding what each agent should do, how we know it works, and what to cut. About half is delivery: planning sprints, unblocking engineers and coordinating launches.
You'll report to the Founder and work daily with a team of engineers who build with tools like Cursor and Claude Code.
What you will do
Own the roadmap for AI CMO's agents, from new channels (influencer, link building, YouTube, LinkedIn) to improving the SEO, GEO, writer, Reddit, and X agents.
Write short, sharp specs: the user problem, what the agent should do, its guardrails, and what good output looks like.
Define success metrics for each agent (traffic, citations in AI answers, drafts approved vs. rejected, activation, retention) and review them weekly.
Work with engineers on agent quality: build eval sets, review real outputs, and triage failures like spammy posts, off-brand copy or made-up facts.
Talk to customers every week and turn feedback, churn reasons and support tickets into prioritized work.
Shape the product as we move upmarket: team workspaces, roles, approval flows and integrations.
Delivery
Run sprint planning, standups and retros with a light process the engineers actually like.
Keep one source of truth for what's in flight, who owns it and when it ships.
Catch dependencies and blockers early, including model providers, third-party APIs and platform rules on Reddit, X and LinkedIn.
Coordinate launches with design, growth and support: release notes, docs, changelog and launch posts.
Run bug and incident triage, and close the loop with affected customers.
What we are looking for
3–6 years in product management, technical program management or engineering, including at least 2 years shipping software at a startup or fast-moving team.
Hands-on experience with LLM products or AI agents. You've written prompts, read model outputs critically, and understand evals, cost, latency and failure modes.
Technical enough to read code, follow a PR, query data with SQL and hold your own in architecture discussions.
Strong product judgment: you cut scope and ship a smaller thing this week rather than a bigger thing next quarter.
Clear, concise writing. People actually read your specs, updates and release notes.
Comfortable with ambiguity and a weekly (or faster) ship cadence.
You use AI tools every day and want to solve distribution for builders like you.
Nice to have
Background in growth, SEO, content or community, or you've grown a product's distribution yourself.
You've built and launched your own product, side project or startup.
Experience taking a self-serve product upmarket to teams.

Work arrangement
Yes