Product Engineer

MakerMaker • San Francisco, CA

1d • Onsite

About The Position

We're building autonomous research agents for recursive self-improvement (multi-agent systems that propose, run, and analyze machine learning experiments). We're a small team based in San Francisco, on-site. Instead of waiting for feature requests, you figure out what the gaps are in the products and systems we have built, and then ship solutions fast. "Users" means two groups: the researchers/engineers running experiments here day to day, and the partner teams we build alongside. The job is the same for both: find the workflow friction that's costing people time, and build the fix. This is build-first. You do the discovery (watch how people work, find the friction, decide what's worth doing) and the building (ship the tool, the interface, the agent, the integration). You own the surface end to end: what to build, how to build it, and whether anyone adopted it. As the work pays off, the role grows into more surfaces and a small team building them with you.

Requirements

• A builder with product taste: you've shipped tools, infrastructure improvements or products that people chose to use, and you can point to the adoption
• Strong generalist engineering: fluent enough across the stack to build the whole thing yourself, in Python plus whatever a surface needs (a backend, a frontend, an agent loop, an integration, serving infrastructure)
• Picking the right problem, not just the assigned one: you've watched users work and changed direction because of it
• Speed with judgment: you ship fast, and you know which corners are safe to cut and which aren't
• Ownership of what gets built, not just how: roadmap, priorities, and trade-offs are yours to make and defend
• Leadership, or readiness for it: you've led or mentored builders, or you're ready to start
• Good written communication: you can make the one-paragraph case for what you're building and why now
• Comfortable around ML: you don't need to be a researcher, but you work with models and agents and pick up the domain fast
• 3+ years building and shipping production software end to end

Nice To Haves

• Early-stage or 0-to-1 product experience, or internal tooling that a technical team genuinely adopted
• Hands-on with LLMs and agent frameworks, or building against model-serving and experimentation infrastructure
• Both engineering depth and a product or founder stint in your background
• Open-source tools that other people use

Responsibilities

• Sit with the researchers, engineers and partners who use our systems and find the workflow friction they've stopped noticing, because they've already worked around it
• Ship the fix (an internal tool, an agent interface, a workflow, an integration, infrastructure improvements), prototyped in days and moved into production once it proves useful
• Own a product surface end to end: decide what's worth building, build it yourself, measure whether anyone adopted it, and cut what they didn't
• Keep the loop short (ship, watch real usage, iterate) rather than committing to a six-month roadmap up front
• Set and defend the priorities for your surface; say no to work that won't move adoption
• Turn a researcher's "I wish the system could…" into something running by the end of the week

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed