About brighthush
brighthush is at the forefront of artificial intelligence, turning cutting‑edge research into products that empower millions worldwide. Our mission is to democratize AI, ensuring that its transformative power is accessible, safe, and aligned with human values. From conversational agents to scalable APIs, we build systems that push the boundaries of what machines can understand and generate, while rigorously prioritizing safety, reliability, and ethical deployment.
As a global leader in AI, brighthush thrives on curiosity, collaboration, and a relentless pursuit of excellence. We foster a culture where bold ideas are encouraged, data drives decisions, and every team member has the opportunity to shape the future of technology. If you’re passionate about building high‑performance, mission‑critical systems that serve real people, brighthush is the place to bring your expertise to life.
Role Overview
We are seeking a seasoned Performance Engineer to join our core engineering organization. In this role, you will be the guardian of speed, reliability, and cost‑efficiency across brighthush’s flagship AI products, including our conversational platform and developer API. You will collaborate with cross‑functional teams—infra, platform, training, and product—to identify bottlenecks, architect scalable solutions, and drive measurable performance gains that directly impact user experience and operational cost.
As an individual contributor, you will own the end‑to‑end performance lifecycle: from profiling and root‑cause analysis to instrumentation, testing, and continuous improvement. Your work will influence architecture decisions, shape system design, and set industry‑leading standards for latency, throughput, and resource utilization.
Key Responsibilities
- Performance Analysis & Optimization – Conduct deep dives into application, middleware, runtime, and infrastructure layers, including networking, storage, Python runtime, GPU utilization, and more. Identify performance regressions and propose actionable improvements.
- Observability & Tooling – Design, develop, and maintain instrumentation, metrics, and dashboards that provide real‑time visibility into system health and performance. Work closely with observability teams to integrate new telemetry streams.
- Cross‑Functional Collaboration – Partner with infra, platform, training, and product teams to set performance goals, define SLAs/SLOs, and align on priorities. Facilitate workshops and design reviews to embed performance thinking early in the development cycle.
- Architecture Influence – Provide expert guidance on system design, data flow, and resource allocation to maximize latency, throughput, and cost‑efficiency at scale. Champion best practices for distributed systems, concurrency, and memory management.
- Root‑Cause Investigation – Lead investigations into high‑impact regressions or scalability incidents in production. Own the post‑mortem process, ensuring root causes are understood, mitigated, and documented.
- Performance Testing & Benchmarking – Define and execute performance testing strategies, including load, stress, and chaos testing. Use results to validate design decisions and refine performance models.
– Drive a culture of continuous performance improvement, encouraging experimentation, data‑driven decision making, and knowledge sharing across the organization.
Essential Qualifications
- 7+ years of software engineering experience with a proven track record in performance or reliability of large‑scale distributed systems.
- Deep expertise in performance profiling tools (e.g., perf, pprof, DTrace, Intel VTune) and tracing systems (e.g., OpenTelemetry, Jaeger, Zipkin).
- Hands‑on experience optimizing across multiple stack layers—database, networking, storage, application runtime, GC tuning, Python or Go internals, GPU utilization.
- Strong understanding of OS internals, scheduling, memory management, and I/O patterns.
- Demonstrated success in building or contributing to observability, benchmarking, or performance‑focused infrastructure at scale.
- Proven ability to navigate ambiguity, influence stakeholders, and drive consensus around performance goals.
- Excellent communication skills—capable of translating complex technical concepts to both technical and non‑technical audiences.
Preferred Qualifications
- Experience with large‑language‑model training pipelines and inference workloads.
- Knowledge of cloud‑native technologies (Kubernetes, Istio, Envoy) and container orchestration at scale.
- Background in high‑performance computing (HPC) or real‑time systems.
- Familiarity with cost‑optimization strategies for GPU and cloud resources.
- Contributor to open‑source performance or observability projects.
Skills & Competencies
- Technical Depth – Mastery of low‑level performance tuning, concurrency, and distributed systems.
- Analytical Thinking – Ability to dissect complex performance issues, synthesize data, and propose elegant solutions.
- Collaboration – Comfortable working with cross‑functional teams, facilitating discussions, and driving alignment.
- Ownership – Take full responsibility for performance outcomes, from hypothesis to implementation to validation.
- Innovation – Constantly seek new tools, techniques, and methodologies to push performance boundaries.
- Communication – Articulate findings, recommendations, and trade‑offs clearly in written and verbal formats.
Career Growth & Learning Opportunities
At brighthush, your growth trajectory is as dynamic as the systems you’ll build. You’ll have the chance to:
- Lead high‑impact performance initiatives that shape the future of AI products.
- Mentor junior engineers, fostering a culture of knowledge sharing and continuous learning.
- Participate in cross‑disciplinary projects, gaining exposure to AI research, product strategy, and infrastructure engineering.
- Attend industry conferences, workshops, and hackathons to stay at the cutting edge of performance engineering.
- Contribute to open‑source projects and internal tooling that benefit the broader engineering community.
Work Environment & Culture
brighthush values a collaborative, inclusive, and transparent work environment. We believe that diverse perspectives drive better solutions, especially when tackling complex AI challenges. Our culture is built on:
- Safety & Responsibility – Prioritizing ethical AI deployment and rigorous safety practices.
- Data‑Driven Decision Making – Leveraging metrics, experiments, and evidence to guide engineering choices.
- Continuous Improvement – Encouraging experimentation, learning from failures, and iterating quickly.
- Work‑Life Balance – Offering flexible schedules, remote work options, and wellness programs.
- Community & Collaboration – Regular team meetups, tech talks, and cross‑team hack days.