Agents
Improve how your team work with individually customised, enterprise-grade AI agents.
Our agents are the core of what we do, we build them from the ground up to the specifications required by enterprise.
Fundamental to how we build our agents is our belief in how to best leverage AI - that it is best used to augment how we work (co-intelligence), not to do our work for us (autonomous intelligence).
With this in mind, we custom build our agents for individuals and their roles and responsibilities. Our agents are jacks of all trades and masters of one. They help improve the quality of their user's work and help them do more, faster.
Fundamental to how we build our agents is our belief in how to best leverage AI - that it is best used to augment how we work (co-intelligence), not to do our work for us (autonomous intelligence).
With this in mind, we custom build our agents for individuals and their roles and responsibilities. Our agents are jacks of all trades and masters of one. They help improve the quality of their user's work and help them do more, faster.
.png)
.png)
AI Agents help us do better work, in less time, and narrow the performance gap between low and high performers.1
+12.2%
Increase in productivity on a selection of tasks
+40%
Increase in quality on a selection of tasks
+25.1%
Increase in speed on a selection of tasks
1 Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality, Harvard Business School, 2023.
Superficial is the only enterprise-grade, artificial co-intelligence platform built on multiple leading models.
We build our agents on a foundation of multiple Large Language Models (LLMs) that are updated regularly and used sequentially to ensure you're getting the most value without being locked into any single vendor.
Sitting atop this is our Security Engine which masks all confidential personal and organisational data to ensure that no private information is shared with underlying LLMs.
From this secure foundation, we construct diverse Agent Operating Systems (Agent OSs) that are trained on the top material from specific functional domains. Users select their specific Agent OS based on their role and responsibilities, and then further train their agents with proprietary organisational and personal data to build contextually aware agents that can truely act as co-intelligence.
Sitting atop this is our Security Engine which masks all confidential personal and organisational data to ensure that no private information is shared with underlying LLMs.
From this secure foundation, we construct diverse Agent Operating Systems (Agent OSs) that are trained on the top material from specific functional domains. Users select their specific Agent OS based on their role and responsibilities, and then further train their agents with proprietary organisational and personal data to build contextually aware agents that can truely act as co-intelligence.
.png)
.png)
Large Language Models (LLMs) are trained primarily on freely available information on the public internet. As a result, their knowledge is both generalised and of varying quality, and therefore of low value to high performing professionals out of the box.
Our agents exceed LLM capabilities through fine-tuning with more individually relevant, higher-quality training data and better contextual awareness of their user and their operating environment.
This additional expert and contextual training atop foundational LLMs enables our agents to act as true co-intelligence to users and produce capabilities that deliver on the promise of AI.
Our agents exceed LLM capabilities through fine-tuning with more individually relevant, higher-quality training data and better contextual awareness of their user and their operating environment.
This additional expert and contextual training atop foundational LLMs enables our agents to act as true co-intelligence to users and produce capabilities that deliver on the promise of AI.
.png)
Co-Creation
Our Agents are highly capable in helping users to plan and review their work with contextual awareness of their individual and organisational styles as well as best practise, role-specific training material.
User Case Study
Assisting a legal associate in drafting a memo in the tone and style of their firm and in the context of the client's regulatory environment and risk appetite.
User Case Study
Assisting a legal associate in drafting a memo in the tone and style of their firm and in the context of the client's regulatory environment and risk appetite.
Decision Making
As humans, the quality of our decision making is limited by our own experiences and biases. Our agents help users improve their decision making quality by leveraging best practice tools and frameworks with the contextual awareness of their user and their operating environment.
User Case Study
Assisting an executive to evaluate the potential impacts and benefits of entering a new market using different mental models and decision making frameworks.
User Case Study
Assisting an executive to evaluate the potential impacts and benefits of entering a new market using different mental models and decision making frameworks.
Work Orchestration
Agents are trained with best practice task management frameworks, helping users to plan and prioritise their work whilst simultaneously learning how to leverage their agent to deliver their best work.
User Case Study
Suggesting optimal times for launching campaigns and organising daily tasks for a marketing manager based on market trends and team availability.
User Case Study
Suggesting optimal times for launching campaigns and organising daily tasks for a marketing manager based on market trends and team availability.
Tactical Coaching
Agents are trained to leverage their expert role-specific training material to provide on-the-job professional and personal coaching for their users.
User Case Study
Assisting a CEO in negotiating a deal using best-practice negotiation strategies whilst understanding contextual awareness of their organisation, competitors and industry.
User Case Study
Assisting a CEO in negotiating a deal using best-practice negotiation strategies whilst understanding contextual awareness of their organisation, competitors and industry.
Ideation
LLMs are naturally strong at ideation and our agents are skilled at helping users to think outside the box with contextual awareness of their organisation and industry, and best practice creative thinking training.
User Case Study
Assisting a growth manager to brainstorm effective growth strategies and role play against customer personas across different markets.
User Case Study
Assisting a growth manager to brainstorm effective growth strategies and role play against customer personas across different markets.
Knowledge Recall
Like all LLM based tools, our agents are exceptionally fast at data synthesis and recall while having the added benefit of deep organisational contextual awareness to get accurate answers quickly, when it matters.
User Case Study
Assisting a sales director in answering a prospect's left-field question during a live sales call using their internal intelligence repository and contextual awareness of their user's organisation and industry.
User Case Study
Assisting a sales director in answering a prospect's left-field question during a live sales call using their internal intelligence repository and contextual awareness of their user's organisation and industry.
Contextual Assistance
Agents are skilled assistants, offering users a rapid way to complete their administrative work and focus on what matters.
User Case Study
Assisting in the summarisation and preparation of meeting minutes in the tone and style of a user's organisation.
User Case Study
Assisting in the summarisation and preparation of meeting minutes in the tone and style of a user's organisation.
Prompt Library
Every user gets access to our community-driven Prompt Library tailored to their role and responsibilities.
Our Prompt Library contains role-specific, best practice prompts to help users get the most out of their agents.
Our Prompt Library contains role-specific, best practice prompts to help users get the most out of their agents.
Thought Logging
A key feature of our agents is their ability to help users capture and explore their thoughts on the fly, helping to connect the dots and find value using their contextual awareness.
Contextual Research
LLMs have vast knowledge but without context, answers are often sub par.
Our agents can assist users in quickly researching and synthesising information as would any LLM but with the contextual awareness of their user and their operating environment to get better quality answers more often.
Our agents can assist users in quickly researching and synthesising information as would any LLM but with the contextual awareness of their user and their operating environment to get better quality answers more often.
.png)