Senior Member of Technical Staff

Blitzy•Cambridge, MA

1d•$180,000 - $225,000•Onsite

About The Position

We are hiring a Senior Member of Technical Staff that has two responsibilities that sharpen each other: take Blitzy into territory it hasn't been tested in real, production-grade open source work at the limits of what autonomous development can do and evaluate the engineers we hire to push it further. Everything you encounter building with Blitzy what holds up, what breaks, what's missing goes directly into the product roadmap. This is the most direct feedback loop we have.

Requirements

Python as primary language; Node.js and JavaScript as supporting
Microservices architecture; REST and gRPC in production
GCP required; deep hands-on experience in at least one of AWS or Azure
Kubernetes at the level where you've debugged production failures, not just deployed workloads
Terraform; infrastructure as code as a discipline, not a convenience
SQL (PostgreSQL, MySQL) and NoSQL (MongoDB, Cassandra, DynamoDB), chosen for the problem, not defaulted to
Graph databases (Neo4j) for complex relational modeling; vector databases for semantic retrieval
LLM-powered systems in production: the full lifecycle, including what happens when models behave unexpectedly at scale
LLM validation loops — evaluation pipelines, regression testing, failure analysis, built and operated, not just designed
LangSmith or equivalent at the depth where you've used it to find real bugs, not just generate traces
Full-stack: sufficient depth to make architectural decisions and debug across the stack end-to-end
Experience designing, integrating, and operating large-scale enterprise systems, including the long-term maintenance tradeoffs that only become visible after the initial build

Responsibilities

Select and build production-grade open source projects using Blitzy's autonomous development platform. The work has to be real enough to expose genuine capability limits, prototypes don't count
Document with precision where the platform holds and where it doesn't, specific failure modes, missing capabilities, incorrect assumptions and translate that into product decisions the engineering team can act on
Hold Blitzy to the standard you would apply to any production system: correctness, reliability, operational durability
Ship open source work that demonstrates what autonomous development can actually produce at the frontier
Assess Principal, Staff, and engineering candidates with the depth that level requires the difference between someone who can reason about a distributed system under failure and someone who has memorized the right things to say about one
Evaluate candidates across the full stack: backend systems, distributed infrastructure, data architecture, LLM/AI systems, and system design under real constraints
Produce structured, specific hiring feedback, the kind that makes the decision obvious, not the kind that defers it
Own the technical bar for engineering hires and evolve it as the platform grows in scope and complexity