Teams at Pramana Labs

Build and evaluate verifiable AI workflows for real-world customer problems. Translate rules and source material into requirements, prototype model-based formalization and symbolic reasoning systems, create rigorous benchmarks, develop applications and integrations, investigate failures, and collaborate with researchers to improve models, proof workflows, and core technology.

Lead research advancing foundation models, formal reasoning, verification, reinforcement learning, synthetic data, and agentic AI. Develop and evaluate training and post-training methods, benchmarks, and structured representations for reliable reasoning. Own end-to-end research initiatives, guide researchers and engineers, and translate breakthroughs into production capabilities. The role requires strong expertise in modern AI research, rigorous experimentation, Python and PyTorch or JAX, plus technical leadership and a record of impactful deployed research.

Build neuro-symbolic AI systems that convert complex knowledge into formal representations and support automated reasoning and verification. Responsibilities include developing model orchestration, evaluation frameworks, data synthesis and post-training pipelines, knowledge representations, program synthesis, and constraint-based inference. The role spans experimentation, benchmarking, failure analysis, optimization, and production-scale implementation, collaborating with AI researchers and formal methods engineers.