AI Assist knowledge companion

Chatbot-like assistant that gives customer-facing representatives real-time access to the knowledge and content they need, right where they work.

Project overview

Spekit AI Assist transforms how teams access critical information, eliminating time-consuming searches and fragmented knowledge bases. This empowers sales teams to focus their energy where it matters most: building meaningful customer relationships and accelerating deal closure. By delivering instant, contextual knowledge access, we're not just streamlining communications—we're fundamentally enhancing sales effectiveness and driving revenue growth

My contributions

Drove end-to-end product development by translating complex ML capabilities into intuitive user experiences—from initial model selection and architecture design through iterative testing, design ideation, user research, and successful beta deployment.

Glossary

FE: Front-end

BE: Back-end

QA: Quality assurance

AI: Artificial intelligence

ML: Machine learning

POC: Proof of concept

KMS: Knowledge management system

Timeline
Lifespan: August 2024 - Present
Beta launch date: October 24th, 2024
The team
x1 Project manager
x1 Senior product designer (that’s me 👋🏻)
x1 Engineering manager
x2 FE engineers
x1 Senior ML Engineer
x1 BE engineer
x2 QA engineers

Problem to be solved

Traditional knowledge management systems are anchored around the 'destination paradigm' - requiring employees to pause their work, identify a need, navigate to a separate system, and search for answers. Research by McKinsey shows that employees spend nearly 20% of their workweek searching for internal information or tracking down colleagues for help. Spekit aims to disrupt this inefficient model by bringing critical knowledge, information, and content directly into employees' natural workflows, eliminating the need to context-switch and ensuring they have instant access to the resources they need, precisely when they need them.

Brainstorming a solution

We explored ways to seamlessly integrate Spekit's AI-powered content recommendations into users' daily workflows. Our solutions included embedding personalized suggestions directly within Gmail's composition interface and developing an intelligent chatbot to provide contextual recommendations in real-time.

Building a proof of concept (POC)

After evaluating multiple solutions, we determined that an AI-powered chat interface would deliver the most value. I collaborated with our engineering team to develop a proof of concept, enabling us to quickly gather user feedback and iterate based on real-world usage. This rapid prototyping approach helped us maintain our competitive edge in the market.

Identifying pain points

After launching the proof of concept, user feedback revealed a fragmented experience across our content discovery features: sidebar recommendations, chatbot assistance, and search functionality were operating as separate solutions to the same core problem. I conducted a comprehensive analysis of these features to identify improvement opportunities and develop a strategy for creating a more cohesive, interconnected knowledge discovery experience.

Gathering inspiration

Through competitive analysis of AI interfaces from Atlassian, Glean, Notion, WalkMe, and Microsoft, I identified successful design patterns and interaction principles. These insights shaped our approach to AI tool design, ensuring both usability and consistency.

Design ideation

To develop a unified user experience across our content discovery features, I crafted multiple design iterations. I gathered feedback from both stakeholders and users to evaluate each design's strengths and weaknesses. This research helped us identify the core requirements for our beta launch while laying the groundwork for our long-term product vision.

External beta program

I designed and managed a strategic beta program with 11 highly engaged customers to validate our product and design hypotheses. I crafted focused pre-beta survey questions to send to users so we could measure specific success metrics and user outcomes.

Feedback analysis

Using Airtable, I transformed the beta feedback into visual charts that clearly communicated user behavior patterns and product opportunities to stakeholders.

Wireframing enhancements based on feedback

By synthesizing feedback from both users and internal teams, I mapped key experience pain points to specific user workflows. These insights drove potential enhancements that I created wireframes for, ensuring each design iteration directly addressed validated user challenges.

Top 3 takeaways

Create clear process documentation

To bridge the communication gap around complex AI and ML systems, it's necessary to create technical documentation supported by intuitive visualizations. This documentation strategy helps streamline cross-team collaboration, facilitate stakeholder alignment, and establish a robust knowledge base for future team scaling.

Keep an eye on the market

In today's rapidly evolving AI landscape, leveraging established design patterns from market leaders while identifying opportunities for meaningful innovation is more important than ever. This approach allows you to build upon proven user interaction models while developing distinctive features that address your unique user needs.

Implement a data driven design approach

This work demonstrated a commitment to data-driven design, moving beyond simple process documentation to focus on strategic impact and measurable outcomes. By maintaining a clear progression from research to implementation, the project successfully balanced user needs with product goals, setting a foundation for future AI feature development.