AI Research Engineer (Agents)

At Coram AI, we’re reimagining video security for the modern world. Our cloud-native platform uses computer vision and AI to help businesses stay safe, make smarter decisions, and move faster; from real-time alerts to seamless clip sharing and multi-site visibility. You’ll be joining a small, fast-moving team that values clarity, craftsmanship, and impact. Every person here has a voice, ships meaningful work, and helps shape how AI can make the world safer and more connected. We are looking for engineers who want to build production-grade AI agents powered by the latest LLMs and Claude Code. This role is focused on turning foundation models into reliable, high-performance systems that operate in real environments. What You’ll Do

• Design and build autonomous agents using state-of-the-art LLMs
• Implement tool use, retrieval pipelines, memory systems, and multi-step reasoning flows
• Engineer prompts and system instructions for robustness, reliability, and speed
• Optimize latency, cost, and throughput in production
• Build evaluation frameworks to measure agent accuracy, tool correctness, and failure modes
• Create high-quality datasets for training, fine-tuning, and benchmarking
• Develop introspection tooling to debug reasoning chains, hallucinations, and tool misuse
• Run structured experiments to improve agent performance through iterative testing

What We’re Looking For

• Strong experimental mindset with a scientific approach to evaluation and iteration
• Experience working with modern LLMs, RAG pipelines, tool calling, and agent frameworks
• Deep understanding of failure modes in LLM systems and how to mitigate them
• Experience building production systems in Python, Go, or TypeScript
• Familiarity with distributed systems, APIs, and real-time infrastructure
• Comfort shipping systems that must be reliable, observable, and measurable

Bonus Points

• Experience building evaluation harnesses or LLM benchmarking systems
• Background in machine learning, applied research, or systems performance optimization
• Experience optimizing inference latency and cost at scale
• Experience debugging complex agent behaviors in real-world environments

Skills and qualifications:

• BS, MS, or PhD in Computer Science, Engineering, Machine Learning, or a related technical field from top University
• 2+ years of experience building software systems (experience working with LLMs, AI agents, or ML systems highly preferred)
• Strong programming ability in Python, with experience in Go or TypeScript a plus
• Experience working with modern LLM APIs (OpenAI, Anthropic, etc.) and building applications powered by foundation models
• Experience building or contributing to production systems that must be reliable, observable, and scalable
• Ability to diagnose and mitigate LLM failure modes such as hallucinations, tool misuse, and reasoning errors
• Strong experimental mindset with a data-driven approach to improving system performance
• Excellent communication skills (written and verbal) in English
• Passion for building cutting-edge AI systems at the speed of a fast-growing startup
• Resilient and adaptable in challenging, fast-paced environments
• Ability to work in an onsite environment, we move faster when we're in the same room

What we offer:

• Competitive compensation package
• 100% Employer-paid medical, dental, vision, and base life insurance
• Flexible paid time off and 9 paid holidays
• 401(k) with both Traditional and Roth options
• Equity in a rapidly growing company
• Referral bonuses
• Daily team dinners and regular team off-sites to build connection and momentum
• The latest Apple tech and unlimited tools so you can win
• Unlimited Cursor and Claude Code credits
• Direct exposure to our AI-native GTM machinery

We're on a mission to transform a $50B+ legacy industry by bringing the power of cutting-edge multimodal LLMs and computer vision to real-world security and operations. From firearm detection to intelligent access control, our AI-native platform turns every camera and sensor into a smart system that enhances safety, efficiency, and awareness. Founded by Ashesh Jain (ex-Lyft Level 5, PhD Cornell) and Peter Ondruska (ex-Lyft, PhD Oxford), Coram AI is backed by Battery Ventures, Mosaic, and 8VC, have raised over $30M, and were named to the CB Insights AI 100 as one of the most promising AI companies in the world. If you're excited to work on mission-critical AI that makes an impact in the real world, we’d love to meet you. Apply tot his job Apply To this Job