All open roles
Engineering
Staff AI Engineer, Consumer Agents
Remote Full-time$260k-$340k + significant equity
You'll lead the modeling work behind every Teleperson consumer agent — the planner that decides which vendor agent to call, the negotiator that pushes back on a denial, and the verifier that double-checks before the user is on the hook. This is core, hard agent engineering: long-horizon tasks, multi-turn tool use, evals you can actually trust.
Why this role matters
This is the role where you stop building chatbots and start building software that does things on behalf of millions of people. The model choices you make here become the default for an entire category.
What you'll do
- Architect the agent loop: planning, sub-agent dispatch, tool selection, memory, and graceful failure
- Stand up an evals stack covering task success, factuality, refusal, and latency budget per consumer task
- Own model selection and prompt-routing across providers (Anthropic, OpenAI, in-house) with clear cost/quality tradeoffs
- Drive the next generation of capability: agents that negotiate refunds, switch providers, and close accounts on the user's behalf
- Mentor the rest of the engineering team on responsible agent design
What we're looking for
- 7+ years engineering, with the last 2+ years building production LLM systems
- Proven track record shipping multi-step agents — not just demos
- Fluent in evals: you have opinions about LLM-as-judge, golden-set sampling, and regression tracking
- Comfortable across Python and TypeScript
- Excited to work fully remote
Nice to have
- Worked on an agent product where the cost of a wrong answer is real money
- Background in RL, planning, or program synthesis
- Published or open-sourced agent infra
Ready to apply?
Send us your resume and a LinkedIn link. We'll get back to you within a week.