Build the Learning Gym, a sandboxed environment for training and evaluating AI agents on real Niural workflows with ground-truth reward signals.
Engineer verifiers, run reinforcement learning loops, and establish honest baselines to prove agent improvements generalize.
Write first-author research papers and internal technical reports, and represent the work externally through preprints and talks.
Niural is a global payroll, Employer of Record (EOR), Agent of Record (AOR), and Contractor Management platform. It is a relentlessly ambitious team focused on building foundational internet infrastructure for the AI era.
Build reinforcement learning and agent environments for real customer and Frontier lab use cases, including task specifications, scoring, and evaluation.
Develop benchmarks and evaluation harnesses to measure model and data quality across accuracy, robustness, safety, latency, and cost.
Run fine-tuning, adapter, and other model experiments to evaluate how data and methods influence model behavior, and deploy local or self-hosted models for evaluation and inference.
Appen has been a leader in AI training data for over 30 years, specializing in human-generated data to train, fine-tune, and evaluate models across generative AI, LLMs, computer vision, and speech recognition. They support model development through an AI-assisted data annotation platform and a global crowd of over 1 million contributors in more than 200 countries, fostering a culture of innovation, collaboration, and humility over ego.
Design and implement state-of-the-art ML models and training pipelines for robotics.
Develop efficient data/training strategies and evaluation frameworks for rapid experimentation.
Collaborate with engineering to optimize training infrastructure and deployment.
We're revolutionizing real-world automation by making robotic systems accessible to everyone. Our AI-powered platform brings software automation to physical spaces, and we're a small startup team working across the stack to solve customer problems.
Build RL environments, agentic systems, LLM pipelines, and evaluation frameworks for real-world AI use cases.
Develop benchmarks and evaluation harnesses to assess model quality across accuracy, safety, latency, and cost.
Conduct fine-tuning and model experiments, deploy self-hosted models, and document reproducible methodologies.
The employer is an organization focused on applied AI research, building practical and reusable AI systems for real-world use cases. It values curiosity, accountability, innovation, collaboration, and continuous learning in a remote environment.
Develop and improve AI methods, algorithms, and systems across the lifecycle of reliable AI agents.
Create new approaches for simulating, evaluating, and optimizing agent behavior in real-world settings.
Turn research ideas into working systems, prototypes, and production-facing capabilities.
The company builds infrastructure that helps enterprises make AI agents more reliable in production. It is a technical and research-focused team working on AI methods and practical systems.
Design, run, and analyse large-scale RL post-training experiments such as RLHF and RL from verifiable rewards.
Research and develop methods that improve out-of-distribution generalization, including reward design, task diversity, and regularization.
Build held-out evaluations to measure generalization and catch reward hacking, and turn successful research into production training recipes.
Cohere is a leading security-first enterprise AI company building cutting-edge foundation models and end-to-end products for businesses. We are a global team of researchers, engineers, and designers passionate about their craft, headquartered in Toronto with offices worldwide.
Develop and improve AI methods, algorithms, and systems across the lifecycle of reliable agents.
Create new approaches for simulating, evaluating, and optimizing agent behavior in real-world settings.
Turn research ideas into working systems, prototypes, and production-facing capabilities.
The company is a research-focused team developing methods and systems to make AI agents more reliable in production. It operates as a small team with a culture of original research, experimentation, and close collaboration with product and engineering.
Build and ship features of AI/ML and LLM-powered systems with guidance from senior engineers.
Implement and maintain AI/ML and AI agent pipelines from data ingestion through model deployment.
Contribute to LLM-powered features such as prompts, evaluations, retrieval, and tool integrations, while documenting experiments clearly.
Robots & Pencils designs AI systems for a human world, pairing engineering with creativity. Teams average fifteen-plus years of experience and value ownership, craft, direct feedback, and continuous learning.
Develop and deploy cutting-edge AI/ML solutions to enhance the platform and improve student learning experiences.
Design, develop, and optimize LLM-powered agentic systems and APIs for real product use cases.
Collaborate with senior engineers and contribute to evaluation frameworks and MLOps pipelines.
Interview Kickstart specializes in interview preparation and career transitions into high-demand tech fields like AI, ML, and Data Science. Over 17,000 tech professionals have been guided by current and former hiring managers to land coveted positions at companies like Google and Amazon.
Lead exploration and application of Large Language Models and Generative AI into core product lines.
Translate the latest research into high-performing systems that enhance customer experiences.
Collaborate with cross-functional teams to set strategy and deploy ML solutions with real business impact.
Replicant is the industry leader in voice AI for customer service, providing automated phone bots for enterprises like AAA and Fanatics. Founded in 2017, the company is backed by top-tier Silicon Valley investors, has offices in the SF Bay Area, Toronto, and New York, and maintains a distributed team across the US and Canada.
Build the component layer around our layout synthesis engine, including API contracts, services, and evaluation gates.
Turn model retraining into a one-command job with built-in benchmarks and readable results for the whole team.
Own latency and cost budgets for learned capabilities and partner with infrastructure engineers on MLOps and deployment.
Higharc is a VC-backed startup that is changing how new homes are designed and built using spatial AI and generative floor plan technology. The company is fully remote, has raised over $175M, and values flexibility, collaboration, and asynchronous deep work.
Develop and train ML models for learned behavior systems using imitation and reinforcement learning.
Write production-quality ML code for training, evaluation, and inference in the autonomy stack.
Collaborate with teams to test and integrate learned behavior models across diverse driving environments.
Torc is an autonomous vehicle technology company that develops software for automated trucks, with a goal to transform freight movement. Now a part of the Daimler family, the company fosters a collaborative, energetic, and team-focused culture.
Design, build, and deploy production ML and LLM-based systems for enterprise clients.
Own technical delivery end-to-end from architecture to deployment and iteration.
Mentor and support other ML engineers through code reviews and technical guidance.
TensorOps is a boutique AI consultancy that designs and ships production-grade AI systems for enterprise clients. The company has shipped AI systems impacting 200M+ end users daily, partnered with 11 unicorns, and operates fully remotely with a supportive, fast-growing culture.
Design, fine-tune, and deploy language model-based solutions from research to production.
Build and improve GenAI pipelines including RAG, corrective RAG, and agentic orchestration.
Develop scalable data and embedding pipelines, and evaluate generative AI outputs against accuracy and safety standards.
Liberty Mutual is a global insurance company that builds generative AI capabilities for its products. It is a large organization with an inclusive culture that values diverse perspectives, flexibility, and professional development.
Build LLM-based agents on the platform's scaffolding, integrating tool calls, internal APIs, and guardrails.
Take agents to production on AWS with containers, CI/CD, secrets, permissions, and security controls.
Define and run evals, monitor with Langfuse, and document runbooks for independent operation.
Muttdata builds innovative Data Products and Machine Learning solutions to help companies solve complex business challenges. It is a fast-growing, remote-first startup that values collaboration, continuous learning, and a positive, ownership-driven culture.
Design, build, and deploy LLM-powered agents that detect spam, harassment, and policy-violating content on Quora.
Develop agentic workflows to automate content moderation and user safety tasks at scale.
Evaluate model performance, investigate failures, and refine AI systems for accuracy and efficiency.
Quora is a global knowledge-sharing platform that connects people with expert answers. This role sits on a small, collaborative engineering team with a remote-first culture and a focus on AI safety and content quality.
Design and improve AI systems for extracting structured reasoning signals from technical interviews.
Build LLM-powered pipelines for semantic analysis, behavioral feature extraction, and evaluation.
Develop scoring, calibration, and benchmarking methods grounded in measurable evidence.
APADCode is an early-stage AI startup building an AI-powered technical interview platform that evaluates how candidates reason and solve problems. The company is a small, early-stage team with a culture of high ownership, direct collaboration, and minimal bureaucracy.
Owns the architectural direction for Code Metal's AI platform and the team building it.
Builds production-grade inference serving, model gateways, agent harnesses, and context-engineering services.
Instruments the stack end to end and drives experimentation and observability for reliable AI systems.
Code Metal provides automated software engineering using AI and formal verification to produce provably correct code. Founded in 2023 and backed by Accel, Salesforce Ventures, and others, it serves defense and enterprise customers from offices in Boston and San Francisco.
Own agent quality end-to-end for Bliro's chat and voice agents, including prompting, context, and model evaluation.
Build eval sets, LLM judges, and benchmarks to measure quality, latency, and cost with statistical rigor.
Establish a repeatable process for improving agent quality and set the long-term technology roadmap.
Bliro builds an AI assistant that handles desk work for field sales teams, with over one million customer touchpoints documented. Backed by leading investors and trusted by German Mittelstand companies, it's a small, high-performing team that values ownership and continuous improvement.
Own AI agents and autonomous workflows powered by LLMs from problem framing through production.
Build RAG pipelines, set evaluation and guardrail standards, and optimize for cost and scalability.
Work directly with client teams to scope problems and deploy AI systems in their environments.
Acceler is a Forward-Deployed Engineer company that takes AI from pilot to production inside client systems. With a practitioner network of nearly 1,000 engineers across the US and India, it's a high-caliber team built on ownership, trust, and rapid learning.