私が開発した最初のB2C共感型AIコンパニオン
月商 — · 創業者1人 · 従業員18人 · Wilmington, Делавэр, США
Aliya Grig（Founder, Evolwe）· 創業 July 2021 · 記事掲載 — · evolwe.ai · 元の記事
創業者は2015年から共感AIの構想を開始。2021年に法人化し、75万ドル（約1.1億円）の資金を投じてプロトタイプ開発とB2C市場の仮説検証を行った。現在はB2Bを重点領域とし、10社の主要クライアントと30社のパイプラインを抱える。B2CはSensEIとして展開し、月間アクティブユーザーは620人（2022年時点）で、有料顧客も獲得済み。従業員は18人。2024年に3.8万件の新規購読、B2Bは12社獲得を目指している。
売値の出典: 記事の本文（2022年テスト時点）（2026-10-13 取得）
If a product is truly in demand, it will attract users without the need for significant marketing expenditure.
Never commit to something that doesn't truly resonate with you… It's crucial to pursue what genuinely drives and inspires you.
広告: Facebook Ads, Instagram Ads, Google Adwords, Google ads
分析: Google Analytics, Google Tag Ma..., Google Forms
ブログ: WordPress
customer service: ManyChat
デザイン: Figma, Canva, Adobe Suite, Figma, Adobe XD
education: Udemy
メール配信: MailChimp, JSON-LD, Linkedin sale...
financing: Shopify Capital
hiring: Upwork
その他: Amazon, AWS infrastru..., WhatsApp, Telegram, Google Sheets, Grammarly, Google Alerts, Typeform, GitHub, Firebase, Loom, Azure, Google Drive, Zoom, Google Hangout, Miro, Lightroom, Google Search..., PostgreSQL, Python, Google Trends, GitLab, Photoshop, Quora
決済: PayPal, Shopify Payments, Stripe, Amazon Payments
ネットショップ: Tilda, Digital Ocean, Amazon Web Se..., Django, firebase
podcast: SoundCloud, Spotify
業務: G Suite, Google Workspace, Slack, Microsoft Off..., Apple Keynote, Notion, Calendly, Discord, Google Meet
reviews: Google My Bus...
発送: Amazon FBA
SNS: Facebook, Instagram, Twitter, YouTube, Pinterest, LinkedIn, Reddit, Medium, TikTok
stock images: Unsplash, Snapseed, InShot
wholesale: Handshake
Born into an astrophysicist family, I was inspired by scientists and artists from childhood, shaping my cosmic vision. Fascinated by the human brain's potential, I envisioned applying this knowledge to AI development, believing it could enhance our understanding of human consciousness and vice versa. I began researching human consciousness in 2010, exploring various philosophical and literary perspectives and interviewing experts. My studies expanded in 2015 to include practical meditation experiments, leveraging Jean Becchio's hypnosis work to deepen my understanding. In 2016, my focus shifted to neuro interfaces, exploring their potential in interpreting brain signals related to consciousness. This led me to conclude that AI development is intrinsically linked to neuroscience, aiming to incorporate human capabilities into AI algorithms. Driven by the ambition to create an AI that not only assists in business but also understands emotions and ideas, I embarked on a mission to develop a technology that could converse and collaborate with humans, enhancing our creative and intellectual endeavors. From 2018, my work involved analyzing AI-human conversations to decipher consciousness. Since 2022, my focus has been on AGI, devising architectural models based on human consciousness studies. The concept for Evolwe AI emerged in 2015, inspired by a discussion with my mentor, igniting the idea of digitizing human experience into AI. This sparked the development of Evolwe AI. Overall, my journey reflects a deep commitment to leveraging AI to improve human life, focusing on creating technology that simplifies tasks and expands access to opportunities for more people.
Evolwe began as a Deep Tech AI project, focusing on emotion recognition and user profiling through text analysis. In 2021, we developed products based on our AI technology and tested various hypotheses. We conducted customer development interviews across different U.S. demographics to understand the potential of an AI companion and identify core product values. Our research indicated a strong demand among young professionals and women aged 25-35 for products aiding in well-being and self-development. These groups, often grappling with stress and anxiety, showed willingness to invest in mental health solutions. We tested 21 marketing hypotheses on a sample of 620 people, utilizing the Jobs-To-Be-Done (JTBD) framework and a lean startup approach. Our AI bot MVPs, catering to diverse needs, were tested on groups of 30-40 people. After analyzing key metrics, we chose to focus on growing SensEI as a personal AI companion, primarily targeting the U.S. and Canadian markets. In September 2022, we joined the FemTechLab accelerator in London, and by November, we were recognized as a top AI startup by NVIDIA. Since then, we've been exploring B2B applications of our API in gaming and the metaverse. Currently, we serve ten core B2B clients and have 30 more in the pipeline. We spent $750k launching the platform, building the prototypes, acquiring the first customers, and testing product hypotheses in the B2C segment. This funding comprised investor money from low-profile sources combined with self-funded contributions. If a product is truly in demand, it will attract users without the need for significant marketing expenditure.
In the beginning, I initiated pre-research on creating an empathetic AI. This endeavor took almost three years, during which I developed the project and hired a team through extensive networking. My passion for this business is shared by the team I assembled, which is fortunate. Subsequently, my CTO and a group of psychologists spent a year annotating conversations to train the AI model in empathy and emotion recognition. We tested the product with about 100 users in each trial. Currently, we have around 620 monthly active users and are planning a significant expansion of our B2C product starting in March 2022. Our launch strategy was detailed and multi-faceted. We focused on creating a unique technological API that facilitates the building and testing of both B2B and B2C products. We adopted a lean approach, testing 4-to 5 product and marketing hypotheses each month. We made decisions based on data and user feedback, continually updating our tactics and processes. We utilized volunteers and interns, more than 450 in total, to train the AI, significantly reducing product development costs. We secured our first paying B2C clients 1.5 years post-launch and entered the B2B market after two years. Our total expenditure amounted to approximately $750k, with a significant portion allocated to technical development, including server and equipment costs, and the remaining to product development and marketing. We allocated a minimal budget to marketing, primarily relying on zero-cost promotional strategies. The most crucial lessons we learned include the importance of testing hypotheses and analyzing market trends in parallel with technical development. It's essential to understand the market dynamics in both B2B and B2C sectors. The a need to focus on building the core of the product rather than hastily implementing each new hypothesis, especially for a deep-tech AI product. The effectiveness of limiting marketing budgets, testing quickly and cost-effectively. If a product is truly in demand, it will attract users without the need for significant marketing expenditure. Overspending on marketing often leads to negligible returns.
In our early stages, we conducted five hypothesis tests to identify our audience segments. Remarkably, 80% of users subscribed, with two hypotheses showing a 100% conversion rate. We learned that beyond appealing marketing metrics, the real value is in a product people are willing to pay for. We now test 2-3 hypotheses monthly, with SensEI's first core audience becoming paying customers. We explored various hypotheses with significant results. For instance, a guideline for stress-relief in women aged 25-35 in demanding jobs showed that 50% felt improved after three days, with 20% willing to pay for continued service. Another test provided coaching for single women seeking partners, resulting in 80% experiencing reduced negative emotions and 40% ready to subscribe at $50/month. We also tested an AI Heart-Break Coach for similar demographics, offering different solutions to two groups. One group received a free session and 50% were willing to pay $25 for a second. The other group, given a three-day guideline, showed a 50% willingness to pay for continued guidance. Another hypothesis tested a self-reflection diary for high-income professionals aged 30-37, resulting in a 40% retention rate after six weeks and a 15% willingness to pay a $15 monthly subscription. Our research indicates that our primary audience is keen on diverse self-improvement resources, frequently accessing them through mid-tier and major influencers on Instagram, such as Deepak Chopra and Joe Dispenza. They also engage with podcasts, YouTube videos, and interviews, and show interest in offline events like retreats, seminars, and festivals. Evolwe’s marketing strategy focuses on young adults interested in self-development, leveraging platforms like Instagram, podcasts, YouTube, and offline events. We utilized cost-effective marketing tools such as SEO, viral product features, directory listings, Instagram growth strategies, and affiliate programs. We plan to collaborate with YouTube and Instagram influencers specializing in well-being, self-development, and psychology, leveraging their large followings and influential status. Our targeted content aims to resonate with their audiences, and we've identified 320 potential influencers in these niches. Our focus includes partnering with micro and mid-size influencers. Further, we're enhancing our deep tech focus, contributing to international conferences, and writing scientific articles and blogs, particularly on Medium. We also use my brand for business development, continuously improve onboarding and user flow, and regularly work on metrics and dashboards to optimize retention, conversion, and engagement. Speed and adaptability in testing different channels and hypotheses are key, ideally within one-week sprints. Never commit to something that doesn't truly resonate with you… It's crucial to pursue what genuinely drives and inspires you.
We have a Roadmap built around the company's mission, vision, and strategy. Also, we have global goals for five years, three years, and a year. On this basis, technical, product, and marketing tasks are decomposed. One of the critical parts of our work is educational meet-ups with the development team, which have become very useful. The non-technical part of our team acquired a deeper understanding of the project. This, in turn, made it easier to develop the product. We have 4 priority values: First, creating the first AI empathetic companion (B2C product) that will be useful for people in daily life - entertain, support, filter social media (help to reduce time and energy), and assist in the most efficient way to accomplish life mission. Second, creating a unique technological solution in AGI combines a personalized approach (emotional and psychological), human understanding, and consciousness that communicates in text and voice format. Third, creating a platform (B2B product) to be used by the largest Web3 and gaming companies to bring more life into games, customize game scripts for players, and educate people through entertainment. Fourth, become the market leader in the field of empathetic AI. We also set short-term goals and tasks. We work on a framework of comparing one goal to another, thus highlighting the one to be chosen. The question may help us decide: “What will we be ready to give up to get another one”?. This framework allows us a lot in setting both goals and tasks. For B2C products, we target 38k new subscriptions in 2024, 223k new subscriptions in 2025, and 930k new subscriptions in 2026. Also, in 2026, we want to increase our retention to 38%, DAU to 90 users, and CR to 16%. Also, we target to reach 1M users in 2025. For B2B products, we target to reach 12 clients in 2024 and 25 clients in 2025. In 2024, we will create profiles, behavior, and voice recognition. In 2025, we are going to develop an AR appearance for assistants.
I've experienced an exciting journey filled with both highs and lows. The first lesson I've learned is to never commit to something that doesn't truly resonate with you, something that doesn’t ignite a spark within and make your heart sing. It's crucial to pursue what genuinely drives and inspires you. The second lesson is the importance of focusing on the overarching goal while also celebrating even the smallest of achievements. Setting a daily target of accomplishing one small step can effectively turn a dream into reality. This approach not only fosters positive thinking but also propels you more efficiently towards your desired outcome. The third lesson involves maintaining faith and hope, especially when faced with challenges. It's important to stay calm and seek creative solutions to problems that arise. Finally, and importantly, inspire those around you through your actions. Among your peers, you'll find like-minded individuals who can form a supportive community around your product, aiding in its promotion and success.
Our team utilizes a range of tools for efficient workflow and communication. These include various Google tools like Meet, Sheets, Gmail, Calendar, Ads, and Tag Manager. We also rely on Slack, Notion, Miro, Zoom, and various Amazon services. We conduct daily calls using Google Meet. For calls with advisers, we prefer Zoom due to its ease of recording and sharing meetings with team members who couldn't join. Google Calendar is invaluable for scheduling meetings and sending reminders. In marketing, Google Tag Manager and Google Analytics play key roles in tracking essential metrics. Miro is extensively used for brainstorming sessions, setting goals, and approving structures and designs for websites or landing pages, as well as product flow. It's also instrumental in our online corporate events. Communication within the team is streamlined through Slack, with various channels for different topics and private chats for direct conversation. Notion is our go-to for task management and document storage, providing an organized platform for all team members. Recently, we've actively started incorporating Amazon's tools. Our servers are hosted on Amazon, and we're exploring additional services to scale our business ideas.
I draw immense inspiration and knowledge from Lex Fridman's and Andrew Huberman's podcasts. Fridman's insights from AI experts greatly align with my project, enhancing our company's strategy and team dynamics. As a neuroscience Ph.D., Huberman's deep dive into the subject offers valuable insights for integrating neuroscience into our AI architecture. Additionally, various influential books have been crucial in refining our team's focus and project approach. “Nudge: Architecture of Choice” reshaped my view on failure and recovery. “Inspired” by Marty Cagan, “The Innovator's Dilemma” by Clayton Christensen, and “Hard Things About Hard Things” by Ben Horowitz have provided key insights into product management, market innovations, and startup challenges. Books like “Trillion Dollar Coach” by Eric Schmidt, “High Output Management” by Andrew S. Grove, “Principles” by Ray Dalio, and “The Great CEO Within” by Matt Mochary have shaped my leadership style. “Thinking in Systems” by Donella H. Meadows and Jordan Peterson’s has honed my systemic thinking and personal development approach. Lastly, Yann LeCun's writings on AI have deepened my understanding of its technical and ethical aspects.
I've observed some common mistakes that people often make: Excessive Perfectionism: There's a tendency to aim for a flawless project and product right from the start. However, this approach can hinder the product's launch. It's better to aim for progress rather than perfection. Lack of Strategic Thinking and Planning: Many people launch products without long-term plans, relying on rash decisions and incoherent actions. Strategic thinking and thorough planning are crucial for success. Team Disconnectedness: A lack of unity within a team can be detrimental. We faced this challenge ourselves but overcame it by ensuring all team members were synchronized and aligned. Inadequate Decision Analysis: Often, project teams fail to consider how decisions in one area (like marketing or product development) can impact other areas (like technical aspects). Understanding the interconnected consequences of decisions is essential. To overcome these challenges, I recommend the following tips: Firstly, fuel Your Work with Enthusiasm. Before starting a project, consider its value to you. Lack of enthusiasm can lead to burnout. Engaging in work that excites you brings energy, resources, and the right people. Secondly, grow Through Mistakes and Failures. Failures and setbacks are natural but can be learning opportunities. After each setback, write down ten things you learned. This practice can turn misfortune into a source of strength. Thirdly, moderate Competitor Analysis. Learn about your competitors occasionally, such as once or twice a month. Constantly monitoring them can lead to unfavorable comparisons and demotivation. Fourthly, embrace Incremental Progress. Focus on progressing your project step by step. It's important to keep moving in any direction, as action creates opportunities and resources. By following these tips, you'll unlock limitless possibilities for turning your dreams into reality.
We currently have one open position. ML Data Scientiest. Details are at the link. For all questions and positions, feel free to contact us at email info@evolwe.ai.
Instagram of Aliya Grig @aliyagalaktika Instagram of Aliya Grig @aliyagrig Twitter of Aliya Grig @AliyaGrig Twitter of Evolwe LinkedIn of Aliya Grig LinkedIn of Evolwe Official website of Aliya Grig Official website of Evolwe Official website of SensEI Medium of Aliya Grig Medium of Evolwe Youtube of Aliya Grig If you have any questions or comments, drop a comment below!