Software Engineer — AI Infra

About Tasklet

Agents that own the work

Tech

We're solving some of the hardest problems in applied AI agents. Some of our core technical challenges include: - building a cloud-native agent runtime that reliably orchestrates long-running, multi-step tasks across thousands of third-party APIs and MCP servers - context engineering that keeps agents sharp and token-efficient while juggling rich tool sets, user memory, and multi-turn conversations - a connections layer that lets agents reason intelligently across every tool a user has wired up - on-the-fly UI generation (we call them "instant apps") so agents can produce interactive frontends when a chat response isn't enough - cloud-based computer use for when agents need a real browser or desktop We're constantly adapting to stay on the frontier. Nearly every line of code here is written with AI, and we're actively working on closing the loop on recursive self-improvement. We use a modern tech stack for our apps & backend.

The role

**Do not apply through this platform. To apply, you must submit a short video by email. See the full application instructions at [https://tasklet.ai/careers](https://tasklet.ai/careers).**[Tasklet](https://tasklet.ai/) is building the AI platform companies use to run their business on agents. A customer describes an outcome in plain English, and Tasklet gets the work done across their tools and systems—immediately or in the cloud 24×7. We’re looking for unusually strong systems engineers to build the core machinery that makes Tasklet’s agents capable, reliable, efficient, and scalable. The work spans our agent harness, context management, filesystem, and execution environment, as well as model serving, databases, Kubernetes, networking, and cloud infrastructure. This role has a deep systems center of gravity, but no fixed boundary. You may move from reasoning about model behavior to runtimes, storage, distributed systems, or production infrastructure in the course of solving one problem. We care more about intelligence, first-principles reasoning, and strong computer-science fundamentals than a conventional AI résumé or exact stack match. We’re open to candidates from **mid-level to Staff+**. Regardless of level, this is a hands-on IC role: everyone at Tasklet personally builds and ships.## Compensation**$225K–$425K base salary**\**0.20%–0.65% equity**Compensation is calibrated to experience and expected impact. Competitive benefits include medical, dental, vision, 401(k), four weeks of PTO, free lunch, and more.### What you’ll do* Build the agent harness, including orchestration, tool execution, state, evaluation, and failure recovery.* Develop context-management, filesystem, memory, storage, and sandboxed-execution systems.* Work on model routing and serving, databases, Kubernetes, networking, queues, and cloud infrastructure.* Make long-running agent workloads faster, safer, more reliable, and more cost-efficient.* Turn fast-moving advances in models and agent systems into robust production capabilities.* Work across the stack and across role boundaries whenever that is the best way to solve the problem.### Who you are* You’re an exceptional problem solver with deep computer-science and systems fundamentals.* You reason from first principles, form precise mental models, and enjoy understanding what is happening all the way down the stack.* You’ve built and operated complex production systems and can debug across application code, runtimes, storage, networking, and infrastructure.* You care about both elegant architecture and real operational outcomes: performance, reliability, safety, and cost.* You learn quickly, operate well with ambiguity, and prefer responsibility over process.* Prior professional AI experience is not required, but genuine depth of engagement is. You already use current models seriously, probe their limits, follow the field closely, and have your own ideas about how AI systems should work. This role is full-time and primarily in person at our office at **1 Post St. in San Francisco**.