Build, deploy, and grow your AI product faster with our expertise.
At Lights On Software, we've partnered with Cohere to support and accelerate their AI offerings, showcasing our deep expertise in software development and Gen AI applications.
Our collaboration has delivered impactful solutions, including the Cohere OSS Toolkit, a skeleton app that streamlines the deployment process for LLM-based RAG applications. What began as the OSS Toolkit has grown into North, Cohere's core product today.
We also integrated Cohere's LLM into Slack, adding powerful features like document search, web research, and thread summarization to transform team communication into a productivity powerhouse.
We crafted over 100 RAG quick start connectors to seamlessly link Cohere's AI with diverse customer data sources.These projects highlight our ability to build scalable, user-focused AI solutions that integrate cutting-edge language models with robust software architecture, empowering businesses to harness AI effortlessly.
And, we're just getting started!
Our team is integrating North into a widely used enterprise communication platform from a large tech company. We'll share more once the integration is live!
"Lights On team has bailed me out several times by being able to hop in and augment our teams with some really solid technical contributions around infrastructure, build environments, projects where we prototyped some AI training frameworks. ... we've had great success with getting skilled folks joining and being productive from day one. "
We've worked with Tenstorrent to develop several important AI visualization tools: TT-NN Visualizer, Model Explorer, and Route UI. These key projects allowed developers using Tenstorrent's hardware to visualize and optimize AI model execution on their cutting-edge hardware, providing critical insights such as model graphs, memory usage, and performance metrics through powerful visualizations and test data management.
From there, our role expanded as we dove into Datastreams (closed-source), an AI-driven observability system. We contributed heavily to its development, enabling real-time monitoring of infrastructure performance and anomaly detection—key features that keep Tenstorrent's systems running smoothly.
We've also pushed boundaries with projects like the AI Superset Assistant, integrating natural language querying into Apache Superset for intuitive, user-friendly data visualization. Meanwhile, our work on CI/CD and DevOps — from setting up pipelines to Docker image releases and automating documentation builds — has streamlined Tenstorrent's operations.
We're excited to continue pushing our AI and software development skills to the next level with Tenstorrent projects!