Senior Product Manager - Software

Efficient Computer•San Jose, CA

5d•$160,000 - $210,000

About The Position

Efficient is developing the world’s most energy-efficient general-purpose computer processor. Efficient’s patented technology uses 100x less energy than state of the art commercially available ultra-low-power processors and is programmable using standard high-level programming languages and AI/ML frameworks. This level of efficiency makes perpetual, pervasive intelligence possible: run AI/ML continuously on a AA battery for 5-10 years. Our platform’s unprecedented level of efficiency enables IoT devices to intelligently capture and curate first-party data to drive the next major computing revolution. The Senior Product Manager - Software owns the strategy and direction for Efficient Computer’s developer platform - including the compiler, SDK, runtime integrations, model enablement stack, and overall developer experience. This role is critical to customer adoption and long-term ecosystem growth, translating Efficient’s hardware capabilities into a scalable, developer-ready platform that enables customers to build, optimize, and deploy applications efficiently. Success in this role requires thinking beyond individual features and approaching the platform as a complete developer ecosystem. The Software PM is responsible for defining product direction, prioritization, customer requirements, and developer experience outcomes, while partnering closely with software engineering, silicon architecture, applications engineering, marketing, and customer-facing teams. This is a highly cross-functional product leadership role focused on strategy, platform evolution, and business impact — not project management or sprint administration. The Software PM defines what should be built, why it matters, how success is measured, and how priorities align with customer and company goals, while engineering teams own implementation and execution.

Requirements

7+ years of product management experience for developer-facing software platforms - SDKs, compilers, runtimes, APIs, ML tooling, or embedded developer ecosystems
Experience shipping tools and platforms used directly by external developers, including operating through the customer-zero phase
Demonstrated ownership of developer adoption, onboarding success, workflow completion, or related developer productivity metrics
Strong understanding of modern ML inference and deployment pipelines (PyTorch, ONNX, ExecuTorch, LiteRT (formerly TensorFlow Lite), TensorRT, or equivalent)
Experience translating customer workflows and technical constraints into actionable product requirements that engineering can build from without ambiguity
Familiarity with software performance tradeoffs involving memory, latency, throughput, power, and hardware utilization
Ability to work effectively with compiler, runtime, systems software, and silicon architecture teams and to influence without authority across all of them
Ability to establish credibility with highly technical developers, compiler engineers, and ML practitioners through clear thinking, technical depth, and strong customer empathy.

Nice To Haves

Experience with compiler toolchains, IRs, and deployment flows (LLVM, MLIR, TOSA, XLA, or equivalent)
Experience with embedded AI deployment pipelines and edge inference workflows
Experience at a semiconductor or AI hardware company building developer tools for a novel architecture
Familiarity with embedded software ecosystems (FreeRTOS, CMSIS-NN, MCU SDK patterns) and Linux-based embedded development environments and cross-compilation workflows
Experience with performance profiling tools, compiler traces, and benchmark-driven optimization workflows
Experience with benchmarking methodologies (MLPerf or equivalent) - including designing, running, and publishing credible results
Experience operating in early-stage environments with incomplete requirements and rapidly evolving architectures
Experience in startup environments with hybrid PM + DevRel responsibilities

Responsibilities

Define the product roadmap for the compiler, SDK, runtime, model conversion and enablement tooling, and developer platform. Set priorities based on customer requirements, competitive positioning, and strategic platform needs.
Lead framework and model enablement strategy across PyTorch, ONNX, ExecuTorch, LiteRT, and future runtimes, including compatibility, validation coverage, performance expectations, and support commitments.
Drive reduction in time-to-first-working-application (TTFWA) across customer segments. Define the metric, baseline it, track it, and own the roadmap items that move it.
Establish SDK governance: surface area, versioning, API stability commitments, and deprecation policy. Define release quality criteria, validation expectations, and backward compatibility standards.
Drive developer confidence through reproducible benchmarking, deterministic tooling behavior, clear diagnostics, and strong documentation.
Participate in customer engagements to understand workflow friction, deployment blockers, and framework requirements. Represent software platform strategy in technical customer discussions and ecosystem partnerships.
Gather and prioritize developer feedback across evaluation, onboarding, optimization, and deployment phases. Maintain structured mechanisms to feed this directly into roadmap decisions.
Partner with Applications Engineering to define the software platform capabilities necessary for production-quality reference applications and customer evaluation flows, including AI inference, vision, audio, and sensor-processing pipelines.
Define requirements for profiling, debugging, and performance analysis workflows that allow developers to understand and optimize application behavior on Efficient hardware.
Balance long-term platform investments against near-term customer enablement and revenue-driving requirements. Understand when the right call is shipping a polished narrow capability vs. a rough broad one.
Influence silicon decisions based on software needs and constraints. Partner with the Hardware PM and architecture team when compiler or runtime requirements have silicon implications.
Establish and own developer success metrics and developer workflow metrics and diagnostics: build success rates, TTFWA, workflow completion rate, and friction points.
Define and own the SDK release cadence, packaging, distribution, and support lifecycle.