Software Engineer (Platform Team)

Job Title: Software Engineer (Platform Team)

Location: SF OR Palo Alto, CA(Onsite)

Job Description:

What we''re looking for

We need a generalist backend engineer with 3+ years of experience (ideally 5-8) in distributed systems who has shipped and maintained production systems at real scale. You should be comfortable reasoning about concurrency, race conditions, and async request patterns, and have a track record of building stable, reliable distributed infrastructure. Bonus points if you have experience with auction systems, exchanges, or marketplace logic.

What you''ll do

• Build and evolve Client''s auction engine and bid management system – the core orchestration layer that allocates GPU capacity to users based on real-time market dynamics
• Take ownership of platform-level distributed systems work including reservation management, capacity allocation algorithms, and pricing logic
• Work closely with the Supply and Consumption teams to ensure GPUs are optimally allocated and that user requests map cleanly to backend orchestration
• Improve stability, resiliency, and operational maturity of the platform layer – including race condition handling, concurrency management, and system reliability
• Contribute to CI/CD pipelines, cloud environment management, and ingress infrastructure that supports the broader engineering organization
• Ship features end-to-end: from design doc through backend API, database migration, feature flag rollout, and monitoring
• Participate in on-call rotations, triage production incidents, and contribute to operational runbooks and automation that reduce toil over time

Tech stack

Python, PostgreSQL, Kubernetes, Docker, AWS, Google Cloud Platform, Terraform

Company Overview

• Client is a platform for consuming AI infrastructure, specifically GPU, in a flexible way.
• Offers a unique consumption model allowing users to bid on spot markets or have flexible reservations.

Role Requirements

• Platform role requires expertise in distributed systems, preferably Python, with strong fundamentals.

Candidate Requirements

• Looking for candidates with 3-10 years of experience, strong in distributed systems and Python.
• Prefer candidates from big tech or VC-backed startups, open to candidates with complex project experience.

Work Arrangement

• Onsite requirement with flexibility between San Francisco and Palo Alto offices.
• Four days a week in the office, everyone must be in Palo Alto on Mondays.

Interview Process

• Initial tech screen followed by face-to-face interviews covering coding, system design, and collaboration.
• Includes an interview with the CEO and reference checks before an offer is made.

Pain Points

• Past hiring has been ad hoc; looking to ramp up hiring without compromising on quality.
• Need to fill gaps and expand the engineering team quickly.

Ideal Candidate Profile

• Strong foundations in distributed systems, ability to handle complex projects, and eagerness to learn.
• Looking for candidates who can influence architecture and decision-making processes.
  
  

Technical Requirements

• Full-stack capabilities with a focus on backend; experience in distributed systems and Python mandatory.
• Experience with billing systems or marketplace products is a plus but not required.

Growth and Culture

• Mithril is in a growth phase, seeking candidates who are energetic and willing to evolve with the company.
• Emphasis on strong collaboration and flexibility within the engineering team.

Seniority

3+ years of experience in backend/distributed systems engineering

Work experience

• Shipped and maintained production backend systems at real scale
• Built distributed systems handling concurrency and race conditions (must have multiple years of experience doing this within complex systems)

• Experience in big tech or a VC-backed startup

Nice to have

Background in marketplace, auction, billing, or exchange systems

Hard skills

• Proficient in Python
• Kubernetes, Docker, IaC, or observability tooling experience

Miscellaneous

• Multiple tenures of

Traits to avoid

• Single-domain specialists unwilling to work across the stack
• Only built prototypes or MVPs, never maintained production systems