Thinking Machines LabSan Francisco, CAFull-timeOn-siteSponsors US visasPosted 25 days ago
$300k–$350k/yr listed by the employer
Above the typical $210k a year for Software Engineer jobs in California (all levels)How we compare
How we compare. Typical is the median yearly pay of 3,390 Software Engineer jobs in California (all levels) that list pay, from the employers' own postings. About means within 5%.
Build and maintain data classification, retention, deletion, and privacy-control systems across internal systems.
Implement isolation guarantees for customer and user data, including Tinker training data and interface conversations.
Run privacy reviews and threat models for data sources, training runs, and product features, then ship controls or tooling.
Translate legal and policy obligations into code, CI checks, and pipeline enforcement.
Summarised from the posting by AI and can be wrong. The full description is the source.
Requirements
Experience
Built and shipped at least one production privacy or data governance system and owned it through operation
Direct experience enforcing privacy or governing data in an AI/ML data pipeline
Chosen pragmatic controls over ideal ones on real systems and can describe what was shipped
Written a technical requirement or control directly from a legal or policy obligation such as GDPR or CCPA
Experience at an AI lab or with ML infrastructure, such as training data curation, multi-tenant fine-tuning platforms, or inference systems handling user data (preferred)
Implemented a privacy-enhancing technique such as deidentification, differential privacy, or secure enclaves in a real system (preferred)
Experience building data lineage, metadata, or policy-engine tooling, or large-scale deletion and retention systems (preferred)
Background in security engineering or direct experience working with a security team on access control implementation (preferred)
Experience as an early engineer building a privacy function and personally building its systems (preferred)
Skills
Written and operated production code in Python or a comparable language
Your match
Sign in and add your resume to see which of this job's requirements it shows.
Saved to your account, encrypted. The AI that reads it doesn't keep it or train on it.
Benefits
Health insurance
Generous health benefits
Dental & vision
Dental and vision benefits
Paid time off
Unlimited PTO
Parental leave
Paid parental leave
Relocation help
Relocation support as needed
Full job description
From Thinking Machines Lab’s careers site
ABOUT THINKING MACHINES
The mission of Thinking Machines is to build AI that extends human will and judgment. We are training frontier models with Inkling, developing Tinker to let people make models their own, and crafting interfaces that broaden human-AI communication. We believe the future worth building is human, and we're hiring people who want to build it.
ABOUT THE ROLE
Show full descriptionShow less
We're looking for an engineer to help manage how data is handled across Thinking Machines: what we collect and acquire, where it is stored, who and what systems can access it, how long we keep it, and how it's used in training, evaluation, and products. You'll help build the systems that let us protect privacy and make definitive statements about our practices from data inventory and classification through access and purpose controls to retention, deletion, and audit. You'll also collaborate closely with researchers and product engineers to answer privacy questions that don’t have textbook answers.
This is a role for someone who knows privacy deeply and treats it as an engineering problem rather than a checklist. The questions we face are rarely settled by policy alone: what it means to honor a deletion request against a trained model; how customers' fine-tuning data on Tinker should be isolated, never commingled, and only used for the customer; when to deidentify data (and how) vs. apply differential privacy to it. We want someone who can weigh privacy against model quality, research velocity, and product usefulness, pick a defensible answer, and then build it.
WHAT YOU'LL DO
Implement and enforce privacy and data governance across internal systems day to day by building systems like data classification tagging, retention jobs, deletion pipelines, and data classifiers and integrating them across systems at Thinking Machines.
Build the isolation guarantees that keep customer and user data separate and protected, including the training data customers bring to Tinker and the conversations people have with our interfaces.
Run privacy reviews and threat models on new data sources, training runs, and product features, and close them out by shipping the control, the check, or the tooling, not only a recommendation.
Translate legal and policy requirements (GDPR, CCPA, contractual commitments) into working code: policy-as-code, CI checks, and pipeline enforcement, built and maintained by you.
Maintain shared libraries and guardrails that make the private path the default, and fix them when they break or fall out of date.
SKILLS AND QUALIFICATIONS
MINIMUM QUALIFICATIONS
Built and shipped at least one production privacy or data governance system (retention/deletion pipeline, policy evaluation engine, data cataloging and/or lineage system, consent system, or similar) and owned it through to operation, not just design.
Written and operated production code in Python or a comparable language, evidenced by a system you can walk through in detail: what it does, how it fails, how you debugged it.
Direct experience with enforcing privacy or governing data in an AI/ML data pipeline (training data, fine-tuning, inference, or evaluation) and can describe a specific privacy risk you identified or mitigated in one of these systems.
Has chosen pragmatic controls over ideal ones on real systems, and can describe what was shipped as a result.
Has written a technical requirement or control directly from a legal or policy obligation (GDPR, CCPA, or similar), and can name the obligation and the control.
PREFERRED QUALIFICATIONS
Experience at an AI lab or with ML infrastructure: training data curation, multi-tenant fine-tuning platforms, or inference systems that handle user data.
Has implemented a privacy-enhancing technique (deidentification, differential privacy, secure enclaves) in a real system, and can describe where it fell short.
Experience building data lineage, metadata, or policy-engine tooling, or large-scale deletion and retention systems.
Background in security engineering, or has worked directly with a security team on access control implementation.
Has been one of the first engineers building out a privacy function, and can point to the specific systems they personally built, not just the program they ran.
LOGISTICS
Location: This role is based in San Francisco, California.
Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $300,000 - $350,000 USD.
Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.