We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers. This 3 to 6 ...
We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers. This 3 to 6 ...
We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers. This 3 to 6 ...
We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers. This 3 to 6 ...
Member of Technical Staff - Mechanistic Interpretability
San Francisco, CA · On-site
$300K - $500K/yr
We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers.
Member of Technical Staff - Mechanistic Interpretability
San Francisco, CA · On-site
$300K - $500K/yr
We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers.
About the Role As a Member of Technical Staff, Mechanistic Interpretability at Radical Numerics, you will study how multimodal genome language models represent, process, and reason about information ...
About the Role As a Member of Technical Staff, Mechanistic Interpretability at Radical Numerics, you will study how multimodal genome language models represent, process, and reason about information ...
We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers.
We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers.
We're focused on mechanistic interpretability, which aims to discover how neural network parameters map to meaningful algorithms. Some useful analogies might be to think of us as trying to do ...
We're focused on mechanistic interpretability, which aims to discover how neural network parameters map to meaningful algorithms. Some useful analogies might be to think of us as trying to do ...
We're focused on mechanistic interpretability, which aims to discover how neural network parameters map to meaningful algorithms. Some useful analogies might be to think of us as trying to do ...
We're focused on mechanistic interpretability, which aims to discover how neural network parameters map to meaningful algorithms. Some useful analogies might be to think of us as trying to do ...
The right candidate has deep grounding in mechanistic interpretability or a closely related field, clear instincts about what questions matter in a security context, and the ability to advance the ...
The right candidate has deep grounding in mechanistic interpretability or a closely related field, clear instincts about what questions matter in a security context, and the ability to advance the ...
Researcher, Interpretability
San Francisco, CA · On-site
$295K - $445K/yr
You will develop and carry out a research plan in mechanistic interpretability, in close collaboration with a highly motivated team. You will play a critical role in helping OpenAI ensure future ...
Researcher, Interpretability
San Francisco, CA · On-site
$295K - $445K/yr
You will develop and carry out a research plan in mechanistic interpretability, in close collaboration with a highly motivated team. You will play a critical role in helping OpenAI ensure future ...
Research Specialist II
Princeton, NJ · On-site
$41K - $60K/yr
The position focuses on mechanistic interpretability of large language models (LLMs) and other deep learning systems, with the goal of elucidating the internal mechanisms these models use to perform ...
Research Specialist II
Princeton, NJ · On-site
$41K - $60K/yr
The position focuses on mechanistic interpretability of large language models (LLMs) and other deep learning systems, with the goal of elucidating the internal mechanisms these models use to perform ...
The position focuses on mechanistic interpretability of large language models (LLMs) and other deep learning systems, with the goal of elucidating the internal mechanisms these models use to perform ...
The position focuses on mechanistic interpretability of large language models (LLMs) and other deep learning systems, with the goal of elucidating the internal mechanisms these models use to perform ...
Research Specialist II
Princeton, NJ · On-site
The position focuses on mechanistic interpretability of large language models (LLMs) and other deep learning systems, with the goal of elucidating the internal mechanisms these models use to perform ...
Research Specialist II
Princeton, NJ · On-site
The position focuses on mechanistic interpretability of large language models (LLMs) and other deep learning systems, with the goal of elucidating the internal mechanisms these models use to perform ...
We believe that a mechanistic understanding is the most robust way to make advanced systems safe. People mean many different things by "interpretability". We're focused on mechanistic ...
We believe that a mechanistic understanding is the most robust way to make advanced systems safe. People mean many different things by "interpretability". We're focused on mechanistic ...
We believe that a mechanistic understanding is the most robust way to make advanced systems safe. People mean many different things by "interpretability". We're focused on mechanistic ...
We believe that a mechanistic understanding is the most robust way to make advanced systems safe. People mean many different things by "interpretability". We're focused on mechanistic ...
You have a strong publication record at top-tier venues (e.g., NeurIPS, ICML, ICLR) with contributions to mechanistic interpretability, representation analysis, probing methods, or model ...
You have a strong publication record at top-tier venues (e.g., NeurIPS, ICML, ICLR) with contributions to mechanistic interpretability, representation analysis, probing methods, or model ...
You have a strong publication record at top-tier venues (e.g., NeurIPS, ICML, ICLR) with contributions to mechanistic interpretability, representation analysis, probing methods, or model ...
Quick apply
You have a strong publication record at top-tier venues (e.g., NeurIPS, ICML, ICLR) with contributions to mechanistic interpretability, representation analysis, probing methods, or model ...
Preferred : • Experience with mechanistic interpretability, probing, or other techniques for understanding model internals. • Familiarity with red-teaming or adversarial evaluation of post ...
Preferred : • Experience with mechanistic interpretability, probing, or other techniques for understanding model internals. • Familiarity with red-teaming or adversarial evaluation of post ...
Preferred : • Experience with mechanistic interpretability, probing, or other techniques for understanding model internals. • Familiarity with red-teaming or adversarial evaluation of post ...
Preferred : • Experience with mechanistic interpretability, probing, or other techniques for understanding model internals. • Familiarity with red-teaming or adversarial evaluation of post ...
... interpretability, mechanistic interpretability, or model internals (sparse autoencoders, feature steering, etc.). Company : Goodfire is an AI research lab using interpretability to turn AI into ...
... interpretability, mechanistic interpretability, or model internals (sparse autoencoders, feature steering, etc.). Company : Goodfire is an AI research lab using interpretability to turn AI into ...
Strong research background in machine learning, computer vision, multimodal AI, robotics, world models, representation learning, model evaluation, or mechanistic interpretability. * Experience ...
Strong research background in machine learning, computer vision, multimodal AI, robotics, world models, representation learning, model evaluation, or mechanistic interpretability. * Experience ...
Mechanistic Interpretability information
See salary details
$31K - $32.8K
13% of jobs
$33.2K is the 25th percentile. Wages below this are outliers.
$32.8K - $34.5K
56% of jobs
$35K is the 75th percentile. Wages above this are outliers.
$34.5K - $36.3K
26% of jobs
$36.3K - $38.1K
1% of jobs
$38.1K - $39.9K
0% of jobs
$39.9K - $41.6K
0% of jobs
$41.6K - $43.4K
0% of jobs
$43.4K - $45.2K
1% of jobs
$45.2K - $47K
1% of jobs
$47K - $48.7K
1% of jobs
$48.7K - $50.5K
1% of jobs
$31K
$36.3K
$50.5K
How much do mechanistic interpretability jobs pay per year?
As of Jul 7, 2026, the average yearly pay for mechanistic interpretability in the United States is $36,260.00, according to ZipRecruiter salary data. Most workers in this role earn between $33,500.00 and $34,000.00 per year, depending on experience, location, and employer.
How to become mech interp researcher?
To become a mechanistic interpretability researcher, typically a strong background in machine learning, deep learning, and programming (e.g., Python) is required. Gaining expertise through advanced degrees such as a master's or Ph.D. in computer science, neuroscience, or related fields, along with experience in analyzing neural networks and using interpretability tools, is essential for this role.
What is the difference between Mechanistic Interpretability vs Data Scientist?
| Aspect | Mechanistic Interpretability | Data Scientist |
|---|---|---|
| Required credentials | Advanced degrees in AI, ML, or related fields | Degree in Data Science, Statistics, or Computer Science |
| Work environment | Research labs, AI development teams | Business, tech companies, consulting firms |
| Industry usage | AI research, model transparency, safety | Data analysis, predictive modeling, insights |
| Search intent | Understanding model internals, interpretability techniques | Data analysis, insights, model building |
Mechanistic Interpretability focuses on understanding how AI models work internally, often requiring deep technical expertise. Data Scientists analyze data to build models and extract insights. While both roles involve data and algorithms, Mechanistic Interpretability is more research-oriented, emphasizing transparency and safety of AI systems, whereas Data Scientists focus on practical data analysis and modeling for business applications.
Is ML a high paying job?
Mechanistic interpretability is a specialized area within machine learning that often requires advanced skills in deep learning, programming, and mathematics. Salaries for machine learning roles vary widely depending on experience, location, and industry, but generally, ML jobs tend to be well-compensated compared to many other tech roles, especially at senior levels or in research positions. Entry-level positions may offer lower salaries, but experienced professionals in this field can earn high six-figure incomes or more.
Which 5 jobs will survive AI?
Mechanistic interpretability is a specialized field within AI research focused on understanding how models work. Jobs in AI safety, research, and development that require deep technical expertise and critical thinking are likely to persist, as they involve tasks that are difficult to automate. Roles emphasizing creativity, complex problem-solving, and human judgment, such as AI ethicists or interdisciplinary researchers, are also expected to remain relevant.
How does mechanistic interpretability work?
Mechanistic interpretability involves analyzing neural networks by examining their internal components, such as neurons and weights, to understand how they process information. It often requires techniques like feature visualization, circuit analysis, and the use of specialized tools to trace decision pathways, helping researchers identify how specific features influence model outputs.
More about Mechanistic Interpretability jobs
What cities are hiring for Mechanistic Interpretability jobs? Cities with the most Mechanistic Interpretability job openings:
What states have the most Mechanistic Interpretability jobs? States with the most job openings for Mechanistic Interpretability jobs include:
What job categories do people searching Mechanistic Interpretability jobs look for? The top searched job categories for Mechanistic Interpretability jobs are:
Full-time
Posted 18 days ago
Job description
About Vmax
Vmax is an applied research lab developing AI capable of open-ended learning. We are building systems to exceed humans in all capacities by optimizing beyond the local maxima of learning from human expertise.
About the role
LLMs are fantastically powerful and there is a rapidly growing corpus of work devoted to understanding their internal representations and computations. We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers.
This 3 to 6 month fellowship is for PhD students or equivalent early-career researchers who want to work at the intersection of mechanistic interpretability and reinforcement learning. You will own a focused research project, work closely with Vmax technical staff, and contribute to research publications.
Responsibilities
Vmax is an applied research lab developing AI capable of open-ended learning. We are building systems to exceed humans in all capacities by optimizing beyond the local maxima of learning from human expertise.
About the role
LLMs are fantastically powerful and there is a rapidly growing corpus of work devoted to understanding their internal representations and computations. We use the tools of mechanistic interpretability to enhance reinforcement learning by generating intrinsic rewards as a supplement or alternative to downstream human-generated verifiers.
This 3 to 6 month fellowship is for PhD students or equivalent early-career researchers who want to work at the intersection of mechanistic interpretability and reinforcement learning. You will own a focused research project, work closely with Vmax technical staff, and contribute to research publications.
Responsibilities
- Develop mechanistic interpretability methods for understanding internal representations, features, circuits, and computations in language models and agents.
- Investigate how model internals can be used to generate intrinsic rewards, auxiliary objectives, diagnostics, or training signals for reinforcement learning.
- Design and run experiments that test whether interpretability-derived signals improve learning, exploration, generalization, robustness, or sample efficiency.
- Compare internally derived rewards against baselines such as human-generated verifiers, reward models, task-level outcome rewards, and standard intrinsic motivation methods.
- Use techniques such as probing, activation analysis, sparse autoencoders, causal interventions, feature attribution, or representation analysis to study model behavior.
- Analyze failure modes, including reward hacking, spurious features, non-causal correlations, objective misspecification, and overfitting to narrow evaluation distributions.
- Build research code, evaluation harnesses, and experimental infrastructure that make results reproducible and useful to the broader team.
- Communicate research progress clearly through written updates, internal presentations, and final project outputs.
- Currently enrolled in a PhD program in machine learning, computer science, artificial intelligence, computational neuroscience, mathematics, or a related technical field. Exceptional candidates with equivalent research experience may also be considered.
- Track record of research excellence or strong research promise, demonstrated through publications, preprints, open-source work, technical projects, competitions, or publicly available artifacts.
- Working understanding of reinforcement learning.
- Familiarity with mechanistic interpretability, representation analysis, or empirical methods for understanding neural networks.
- Strong programming ability in Python and experience with at least one major ML framework such as PyTorch or JAX.
- Clear written and verbal communication of technical ideas.
- Experience with LLM post-training methods
- Familiarity with intrinsic motivation, unsupervised RL, auxiliary objectives, representation learning for RL, or curiosity-driven learning.
- Experience with scalable ML experimentation, distributed training, experiment tracking, or reproducible research infrastructure.
- Interest in turning mechanistic understanding into practical training methods, rather than only analyzing models after training.
- This role is based in our San Francisco office; for exceptional candidates we are willing to consider a hybrid arrangement