A Note on applying to Pangram:
At Pangram, we firmly believe that genuine human expression, and our ability to communicate with each other online and through the written word, is a fundamental right- and a value worth preserving.
Please don't use AI to answer any of the questions as part of your application - we want to hear from you!
Pangram Labs’ mission is to protect authenticity and preserve human connection in the context of the generative AI era. We build the world’s best AI detection systems. We publish research on state-of-the-art AI detection techniques and algorithms and build products to bring our technology into people's daily lives. We provide tools for individuals, institutions and companies to detect the use of AI in written content.
Pangram is a recognized thought leader in the space of AI detection. Our CEO & Co-founder Max Spero has been named as one of Time Magazine’s 100 Most Influential People in AI and Pangram has been named one of Time’s Best Inventions of 2026.
We are looking for high agency, mission-driven, and passionate builders to join our team!
About the RoleAs a Machine Learning Engineer at Pangram Labs, you will build software to support the machine learning development cycle from data generation, to training models, to deployment and monitoring production machine learning systems in real customer environments. At Pangram, ML engineers are also highly involved in our research effort - in publishing research and regularly contribute ideas and innovations to the team. However, formal research experience is not necessary.
This is a full-time, on-site role in our Brooklyn office reporting to our CTO.
What You'll DoBuild robust data pipelines that mine the Internet at scale and generate millions of synthetic text examples for training detection models
Manage distributed infrastructure for multi-GPU LLM training
Profiling and optimizing training and inference code
Deploy efficient inference pipelines for serving LLMs at scale
You are a great candidate for this role if you have:
B.S./M.S. in Computer Science or related areas
Strong programming skills in Python and modern ML frameworks
Excellent understanding of transformers and LLM fundamentals
Comfort working across research and engineering boundaries
You are an exceptional candidate for this role if you also have:
Experience with NVIDIA GPU programming and CUDA
Experience with distributed training frameworks, such as DeepSpeed, FSDL, Ray
Experience with inference frameworks like vLLM
Experience with large-scale data processing (Spark, Beam) and orchestration (Airflow)
Experience with MLOps and experiment tracking
Experience with DevOps tools
Familiarity with cloud-based infrastructure (AWS/GCP)
Skills Required
- Bachelor’s or master’s degree in Computer Science or a related field
- Strong programming skills in Python and modern machine learning frameworks
- Excellent understanding of transformers and large language model fundamentals
- Comfort working across research and engineering boundaries
- Experience with NVIDIA GPU programming and CUDA
- Experience with distributed training frameworks such as DeepSpeed, FSDL, or Ray
- Experience with inference frameworks such as vLLM
- Experience with large-scale data processing using Spark or Beam
- Experience with orchestration using Airflow
- Experience with MLOps and experiment tracking
- Experience with DevOps tools
- Familiarity with cloud-based infrastructure such as AWS or GCP
What We Do
Pangram's mission is to promote authenticity and preserve human connection in the context of the generative AI era. We firmly believe that genuine human expression, and our ability to communicate with each other online and through the written word, is a fundamental right- and a value worth preserving. At the core, Pangram is best-in-class AI detection software that can be applied to multiple use cases across many verticals. Pangram's core value proposition is its accuracy: a testament to the depth of the research that we have conducted that has been validated by multiple academic third party studies. AI detectors have been widely maligned by the media and the educational community, and we believe somewhat rightfully so: early AI detectors like GPTZero, Copyleaks, ZeroGPT, Turnitin, and had false positive rates of upwards of 1% (false accusations of human writing as AI), and false negatives (missing AI generated text, predicting it as human). They faced many of the issues that still plague AI detection discourse today, such as predicting that the Declaration of Independence is AI-generated and demonstrating elevated false positive rates on ESL writing. We are fighting to earn the trust of the public back by doing AI detection the right way. Public research is core to our identity. We are transparent about what we do, and put our methods out into the open because we have nothing to hide, and we are serious about building trust with our customers. We still believe we are the only AI detector that "actually works," and the only serious team that is up for solving this challenging research problem.
Gallery
