Software Engineer - Training Infrastructure at Baseten (San Francisco, CA)

Baseten· San Francisco, CA· $165k–$330k·3d ago

Role details

Salary

$165k–$330k

Work type

Hybrid

Employment

Full-Time

Equity

Yes

Skills

GoKubernetesDistributed systemsPerformance tuningObservability systemsCloud providers (AWS, GCP)PythonAWSGCPCrusoeDigitalOceanNebius

Benefits

Competitive Compensation · Equity · 100% Coverage Of Medical, Dental, And Vision Insurance · Flexible PTO Policy · Company Wide Winter Break · Paid Parental Leave · Fertility And Family-Building Stipend · Company-Facilitated 401(K)

Baseten at a glance

AI inference platform for deploying, optimizing, and running machine learning models in production at scale.

Baseten runs trained AI models in production for other companies, handling the GPUs, autoscaling, runtime, and performance tuning so engineering teams get a fast, reliable API without operating the infrastructure themselves. It supports open source, custom, and fine tuned models across managed cloud, hybrid, and self hosted deployments.

$2B+ raised · latest: Series F · $1.5B · June 2026 (valuations of $13B and $11B across two tranches) · backed by Altimeter Capital, Conviction, Spark Capital, Sands Capital

Summary

Architect and lead the development of the training platform's infrastructure, including scheduling, storage, networking, and observability. Partner with researchers and developers to translate complex training requirements into scalable technical solutions using Go and Kubernetes.

Job description

ABOUT BASETEN

Baseten powers mission-critical inference for the world's most dynamic AI companies, like Cursor, Notion, OpenEvidence, Abridge, Clay, Gamma, and Writer. By uniting applied AI research, flexible infrastructure, and seamless developer tooling, we enable companies operating at the frontier of AI to bring cutting-edge models into production. We're growing quickly and recently raised our $1.5B Series F, led by Altimeter Capital, Conviction Partners, and Spark Capital. Join us and help build the platform engineers turn to ship AI products.

THE ROLE

As a Software Engineer on the Training Infrastructure team, you'll architect and lead development of our training platform, supporting top tier research engineers and model developers. You'll make key technical decisions for the infrastructure enabling developers to deploy, scale, and monitor their workloads with high performance and reliability. You’ll own scheduling, storage, networking, reliability, and observability of technical systems in the training stack

EXAMPLE INITIATIVES

Take a look at what we’ve built so far:

RESPONSIBILITIES

Design and architect scalable infrastructure systems for our ML training platform (e.g. scheduling, storage, and networking)
Partner closely with developers and research engineers to translate complex training requirements into technical solutions
Design and architect a global training scheduler
Design and architect reinforcement learning systems and continuous learning pipelines
Drive long-term improvements to increase reliability of systems and velocity of development
Partner closely with SRE and Capacity teams to unlock state of the art training infrastructure
Make critical architectural decisions balancing performance with system reliability
Lead technical discussions and mentor junior engineers on infrastructure best practices
Contribute to long-term technical strategy and infrastructure roadmap

REQUIREMENTS

Bachelor’s degree or higher in Computer Science or related field
5+ years of experience
Proficiency in Go, with
Deep expertise with Kubernetes in production environments
Advanced understanding of distributed systems concepts and performance tuning
Proven experience designing observability systems
Experience with ML/AI workloads and MLOps platforms

NICE TO HAVE

Experience with distributed storage systems
Python experience a plus
Extensive experience with major cloud providers (AWS, GCP) and neo-cloud providers (Crusoe, DigitalOcean, Nebius)
Experience with workload orchestration platforms like Temporal or Airflow
Familiarity or experience with the open source training stack and frameworks (NCCL, PyTorch, Megatron, NemoRL, VeRL, Axolotl, HF Trainer) and distributed training techniques (FSDP, DeepSpeed).
Experience developing AI products, tooling, or agents

BENEFITS

Competitive compensation, including meaningful equity
(U.S. only) 100% coverage of medical, dental, and vision insurance for employee and dependents
Flexible PTO policy including company wide Winter Break (our offices are closed from Christmas Eve to New Year's Day!)
Paid parental leave
Fertility and family-building stipend through Carrot
(U.S. only) Company-facilitated 401(k)
Exposure to a variety of ML startups, offering unparalleled learning and networking opportunities.

Apply now to embark on a rewarding journey in shaping the future of AI! If you are a motivated individual with a passion for machine learning and a desire to be part of a collaborative and forward-thinking team, we would love to hear from you. At Baseten, we are committed to fostering a diverse and inclusive workplace. We provide equal employment opportunities to all employees and applicants without regard to race, color, religion, gender, sexual orientation, gender identity or expression, national origin, age, genetic information, disability, or veteran status. We are an Equal Opportunity Employer and will consider qualified applicants with criminal histories in a manner consistent with applicable law (by example, the requirements of the San Francisco Fair Chance Ordinance, where applicable).

Why work at Baseten

Hard Technical Problems: Engineers work on the most challenging problems in modern infrastructure—model serving, low-level GPU optimization, networking, distributed systems, and observability. This is a high-agency, high-impact engineering environment.
High Growth Trajectory: The company is experiencing explosive growth (224% headcount increase, active hiring). This offers significant career acceleration and ownership opportunities.
Strong Engineering Culture: Founded by engineers, for engineers. The culture emphasizes "first-principles thinking across the entire stack" and a "customer-obsessed" mindset. The employer rating on compensation, culture, and work-life balance is rated highly (5.0).
Top-Tier Team & Investors: The team has strong talent density with hires from Meta, Stripe, Google, NVIDIA, and Databricks. Being backed by top-tier VCs provides stability and a clear long-term vision.
Hybrid/In-Office: Based in San Francisco with a strong in-person or hybrid culture common for fast-moving infrastructure startups.

About Baseten

Company Overview

One-liner: Baseten is an AI infrastructure platform that provides high-performance inference infrastructure for deploying, optimizing, and managing AI models in production.
Entity Type: Private (Series F)
Headquarters: San Francisco, California, United States
Founded: 2019
Founders: Tuhin Srivastava (CEO), Amir Haghighat (CTO), Phil Howes (Co-Founder & Chief Scientist)

Core Business

Primary industry/industries: AI Infrastructure / Cloud Computing / MLOps
Target customers: B2B, targeting AI-native companies, enterprises, and developers building and deploying large-scale AI applications. Notable customers include Abridge, Cursor, Lovable, Notion, and OpenEvidence.
Mission or purpose statement: Provide the tooling, expertise, and hardware needed for teams to bring performant AI products to market as quickly and reliably as possible. The company's core belief is that "inference is everything" and that the future of AI will consist of millions of specialized models embedded into every product and workflow.

Products & Services

Baseten Cloud: A fully-managed, global deployment platform for AI models. Provides out-of-the-box performance optimizations, 99.99% uptime, and massive horizontal scale. Supports single-tenant and self-hosted (hybrid) deployments.
Baseten Inference Stack: The proprietary, underlying technology that powers the platform. Includes custom model runtimes (LLMs, embeddings, voice AI, image generation), advanced caching, custom kernels, and cutting-edge performance research. Claims 2x higher throughput and 10% lower latency for embeddings (BEI).
Baseten Chains: Enables granular hardware and autoscaling for compound AI systems, reportedly powering 6x better GPU usage and cutting latency in half.
Frontier Gateway (Pre-optimized Model APIs): Allows users to instantly test and deploy pre-optimized versions of the latest open-source, custom, and fine-tuned models (e.g., Qwen, DeepSeek, GLM, gpt-oss).
Forward Deployed Engineering (FDE): A service offering where Baseten engineers work side-by-side with key customers to build, optimize, and scale their models from prototype to production.
Training on Baseten: Allows users to train models and deploy them in one click on the inference-optimized infrastructure.
Monetization API: An inference API that allows model developers to monetize their models.

Market Standing

Valuation/Market Cap: Not privately disclosed, but the company has raised a total of $585M in funding.
Key Metric: Annual revenue of $7.0M (as per LinkedIn). Total funding of $585M.
Notable Investors/Partners: IVP, Spark Capital, Bond, CapitalG, South Park Commons, Greylock.
Growth Signals:
- Announced a Series F funding round (details not fully public in search results).
- Massive headcount growth: 224.4% year-over-year, now employing 201 people.
- LinkedIn follower growth: 199.9% year-over-year.
- Active job postings: 112 open roles.
- Acquired the company Parsed.
- Backed by high-profile institutional investors including CapitalG (Alphabet's growth fund) and IVP.

Competitive Advantages

High-Performance Inference Stack: A proprietary, deeply optimized stack focusing on ultra-low latency and high throughput for the most demanding Gen AI workloads (voice AI, real-time streaming, complex compound AI systems).
Customer Obsession & Forward Deployed Engineering: A hands-on, consulting-style approach with their FDE team, which builds deep, sticky relationships with high-value customers like Abridge and Notion.
Global & Hybrid Cloud Flexibility: Offers both fully-managed cloud and self-hosted deployments within customer VPCs, providing scalability and security for enterprises with strict data governance requirements.
Founding Team's Deep Technical Credibility: Built by former researchers and engineers who experienced the "deployment hell" of AI systems firsthand, giving them deep empathy and expertise in the problem they are solving.

Strategic Focus

Dominating the "Inference-as-a-Service" Market: The company's core thesis is that inference will become the largest market ever created, and they are building the infrastructure to power it.
Scaling for High-Volume, Mission-Critical Workloads: Focus on reliability, performance, and cost-efficiency for customers who cannot afford downtime or latency (e.g., AI phone calls, real-time translation).
Deepening the Technical Stack: Continuously investing in performance research (custom kernels, new decoding techniques) and expanding capabilities into new areas like voice AI and audio streaming.
Aggressive Hiring: With 112 open positions and massive YoY growth, scaling the engineering, sales, and operations teams is a clear priority.

Sources