Member of the Technical Staff - Chatbot Engineer

Two Dots • San Francisco, CA

4d • $175,000 - $275,000

About The Position

Two Dots is seeking a software engineer to build consumer-facing chat agents that serve as the frontend to complex workflows. This role requires a blend of user empathy, strong written English, Python proficiency, and a metrics-driven approach. The ideal candidate will be comfortable using SQL or BigQuery for quality analysis and possess the ability to perform manual QA when necessary. This position is akin to a future UX Engineer, but focused on conversational natural language experiences.

Requirements

• Deep understanding of context management, including the distinction between workflow LLM calls and true agent loops with tool calling.
• Ability to transition from smart models to cheaper, faster ones without relying on prompt hacks or extensive rule lists.
• Understanding of the boundary between tools/APIs and natural language.
• Ability to differentiate between structural elements and those related to tone, framing, or model "dark magic."
• Focus on the model's operational headspace, user experience quality, and product functionality for real users.
• Avoidance of "Gas Town" mentality; not believing every problem requires a meta-harness or outsourcing judgment to chatbots.
• Knowledge of when to escalate issues to MLEs for fine-tuning or advanced methods.
• Deep care for user outcomes, including measuring experiment performance and proactively solving quality issues.
• High frustration tolerance for ambiguous chatbot engineering tasks.
• Proficiency in Python is preferred; TypeScript or other strong software engineering backgrounds are also acceptable.
• Strong programming ability to build reliable systems manually, not solely relying on AI coding tools.

Nice To Haves

• Meaningful professional or personal experience building chat agents that interact with real systems.

Responsibilities

• Build consumer-facing chatbots that act as the frontend for complex workflows.
• Integrate internal workflow APIs and domain object code with the real-world interaction patterns of AI agents.
• Optimize smaller models to perform at the level of larger models.
• Develop innovative methods for automating product judgment, such as using chatbots for user role-playing instead of solely relying on manual QA or fixed test cases.
• Collaborate closely with design and product teams to balance user interface aesthetics, interaction quality, and business objectives.

What This Job Offers

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