Requests for startups in 2025

“Products that are loved and used by customers don’t start with what you want to build; they start with what your customers need.”

– Des Traynor, co-founder of Intercom

AI startups attracted a record 46.4% of the total $209 billion raised in venture capital funding in 2024. This signifies a substantial increase from less than 10% a decade earlier. According to the CB Insights 2024 State of Venture Capital report, while VC activity somewhat recovered from the low point of 2023, deal volume was hit an 8 year low in 2024. It seems that it’s more important than ever to be thoughtful about what you’re building.

Requests for startups are a way for investors to communicate what might work and where they are actively looking for startup teams. It’s not perfect of course, and the majority of opportunities will come from entrepreneurs and not from investors, but it’s a good barometer of the current Zeitgeist in tech. This is a follow up to a similar post I published in 2024.

Now is the best time in history to be a builder. The golden age of innovation is here, and I’ve compiled a list of exciting startup ideas for 2025, inspired by the latest trends from A16Z, Y Combinator, Stripe, Conviction and other thought leaders. Let’s dive in.

Ycombinator’s RFS 2025

Ycombinator updated its Requests for startups list in December 2025, and while it’s very US centric (with a wink to a Trump administration and crypto friendly regulation) it captures the current zeitgeist.

1. Government software

• AI software to automate government administrative tasks
• Target large-scale operations like form processing, application reviews
• Build solutions for citizen-facing services (DMV, permits, licenses)

2. Public safety technology

• Advanced computer vision: develop privacy-preserving surveillance
• Emergency response systems: cut response times and improve coordination
• Community safety platforms: bridge community-police communication gap
• Law enforcement efficiency tools: reduce paperwork, increase patrol time

3. US manufacturing

• Focus on automated manufacturing systems using ML and robotics
• Follow Tesla’s Gigafactory model for modern US manufacturing
• Target industries with strategic importance (semiconductors, batteries)

4. Stablecoins 2.0

• Build business-focused stablecoin management platforms
• Develop easy integration tools for developers
• Focus on cross-border payments and remittances

5. LLMs for chip design

• ASIC and FPGA optimisation tools using AI
• Target 5–100x performance improvements
• Focus on energy efficiency (10–100x gains)

6. Fintech 2.0

• Build AI-first financial products without legacy constraints
• Target underserved markets in insurance and wealth management
• Focus on global payment solutions

7. Space companies

• Capitalise on 10x launch cost reduction
• Build businesses around satellite applications
• Create infrastructure for routine space access

8. AI-aided engineering tools

• Modernise CAD/CAM with AI assistance
• Simplify complex engineering tools
• Create accessible design validation tools, focus on physics simulation acceleration

9. One million jobs 2.0

• Create sustainable human-centric job platforms, focus on jobs that AI can’t replace
• Build tools for independent business owners

A16Z big ideas in Tech 2025

Andreessen Horowitz (A16z) suggests that the big ideas in tech for 2025 are driven by advancements in AI, evolving consumer behaviours, and the increasing convergence of technology with various industries. The predictions offer significant opportunities for startups across different sectors. Here are some key areas and related startup opportunities:

AI for the Physical World

• The resurgence of nuclear energy: A16z predicts a surge in demand for nuclear energy, driven by regulatory reform, public enthusiasm, and the growing energy needs of AI data centres. This presents opportunities for startups to develop technologies that support the safe and efficient operation of nuclear power plants, including advanced reactor designs, waste management solutions, and grid integration systems.
• Jobs crossing the hardware-software chasm: The rise of AI in hardware contexts is creating a demand for engineers and technicians skilled in both hardware and software. Startups can capitalise on this by providing training and development programs for these specialised roles, or by building tools and platforms that bridge the gap between hardware and software development.
• Decentralised defence: A16z anticipates increased reliance on autonomous systems and AI in military operations, requiring robust edge computing and power solutions in challenging environments. Startups can develop specialised hardware and software for decentralised defence applications, focusing on areas like ruggedised edge computing devices, energy-efficient power sources for remote deployments, and secure communication technologies for autonomous systems.

AI A,ugmenting Human Capabilities

• AI brain: A16z predicts the rise of personalised “digital brains” powered by AI, leveraging personal digital data to help individuals understand themselves better, communicate more effectively, and improve productivity. Startups can develop applications that build upon this concept, offering features like personalised insights, AI-assisted communication tools, and AI-powered productivity enhancements. This aligns with our previous discussion about Ark Invest’s prediction of AI lowering daily work hours by improving productivity.
• Real-time AI: The emergence of real-time AI is transforming creative workflows and enabling novel applications like AI bandmates and personalised education. Startups can focus on developing real-time AI tools for specific creative domains, such as music production, video editing, or interactive storytelling.
• Hyper-personalised knowledge work: A16z expects AI tools to become more personalised, capturing individual styles and preferences in knowledge work tasks like writing and presentation creation. Startups can develop AI-powered tools that enable personalised content creation, offering features like style adaptation, tone matching, and custom formatting.

AI-Driven Business Transformation

• Regulation as code: A16z sees an opportunity for startups to use LLMs to simplify compliance by turning complex regulations into easily understandable queries. This aligns with our previous discussion on the potential for AI-aided engineering tools to automate knowledge work and improve efficiency. Startups could build platforms that translate regulations into code, enabling businesses to automate compliance processes and reduce the burden of staying up to date on regulatory changes.
• Ripping out legacy systems: A16z predicts that AI is prompting enterprise buyers to replace legacy systems of record with dynamic, AI-powered solutions. Startups can capitalise on this trend by developing modern, AI-native systems of engagement that offer superior functionality, flexibility, and user experience compared to traditional systems.
• Super staffing in healthcare: A16z believes AI will serve as a “super staffing” platform in healthcare, addressing the staffing crisis and automating administrative tasks. Startups can focus on building AI-powered solutions that augment healthcare workers, streamline administrative processes, and improve patient care.

Democratising Access and Ownership

• Democratising health through tech: A16z emphasises the role of technology in empowering patients to take control of their health. Startups can develop accessible and user-friendly tools for personal health monitoring, data analysis, and personalised health recommendations.
• Tokenising unconventional assets: A16z predicts a wider adoption of tokenisation across various sectors, enabling liquidity for previously inaccessible assets. Startups can build platforms and services that facilitate the tokenisation of diverse assets, creating new opportunities for investment, ownership, and participation in the digital economy.

A16z’s insights into the big ideas for 2025 emphasise the transformative power of AI and its potential to create new opportunities for startups across a wide range of industries. By embracing these trends and developing innovative solutions, startups can play a pivotal role in shaping the future of technology and its impact on society.

Sarah Guo and Conviction Capital

Sarah Guo is the founder of Conviction, an early stage VC fund, Embed is an acceleration program launched by Conviction, investing $150K into pre-seed AI startups. Conviction believes that the abundance of AI technology will permeate every part of the economy, creating opportunities for startups in both traditional and emerging sectors. They highlight several areas ripe for disruption and offer suggestions for how startups can capitalize on these trends. Many of these ideas align with the predictions we discussed from Y Combinator, Ark Invest, and Andreessen Horowitz.

Here are some of the big ideas in tech for 2025 according to Conviction, with a focus on startup opportunities:

AI for the Physical World:

• The New Datacenter: The shift in datacenter workloads towards AI, particularly with the dominance of Nvidia GPUs, presents opportunities for startups to develop alternative chip architectures, optimised for AI workloads like transformers and matrix math. Conviction suggests exploring options like chips reconfigured for reduced bandwidth requirements, alternatives to NVLink, full-stack optimised systems for specific workloads, and latency-bound inference solutions. They also highlight the demand for a second source to Nvidia, creating opportunities for partnerships with companies like AMD.
• Understanding and Generating the (3D) World: Despite advancements in foundation models, generating high-fidelity 3D models remains challenging. Conviction sees opportunities for startups to develop solutions for generating 3D assets for non-toy applications, particularly in areas like entertainment, education, and commerce. Research into multimodal learning to infer 3D data from 2D sources and efficient data generation techniques will be crucial. They also see a longer-term opportunity in tackling the “high precision” problem of generating 3D assets suitable for demanding applications like construction and manufacturing. Startups can explore assistive tooling that helps engineers narrow down their design space or perform menial tasks. They also suggest combining generative models with simulation and post-processing techniques to improve the quality of generated outputs.

AI Augmenting Human Capabilities:

• Finally, Manageable Metadata: Conviction believes that AI can revolutionise data cataloging and metadata collection, a traditionally manual and painful process. They see an opportunity for startups to develop AI-powered solutions that automate metadata generation, identify data quality issues, apply policies, and extract context from user interactions. These solutions will disrupt legacy data catalogs and become a keystone in the AI data stack.
• Next-Gen Autocomplete: Conviction sees an opportunity to create a “Copilot-like” experience for various computing tasks, particularly email and other authoring activities. Startups can develop AI-powered browser extensions that learn a user’s writing style and provide personalised autocomplete suggestions, making them more efficient. They suggest leveraging multimodality capabilities to create a “second brain” that can understand various input forms like screenshots and voice notes. Conviction believes that incumbents will be slow to adopt this technology, creating space for new entrants. They emphasise the importance of clever consumer distribution, speed, and privacy, suggesting a hybrid local/cloud approach.

AI-Driven Business Transformation

• Ownership is the New Sales: Conviction challenges the traditional approach of selling AI tools to fragmented industries, arguing that it might be more effective to own and improve these businesses top-down. They see opportunities for startups to acquire and operate businesses in sectors like HOAs, BPOs, accounting, dev shops, and legacy software, using AI-based automation to drive value.
• Automated Root Cause Analysis: Conviction believes that AI can significantly improve incident resolution times and the on-call experience for engineers. Startups can develop AI agents that analyse logs and metrics to identify root causes, suggest fixes, and generate post-mortems. They suggest leveraging existing runbooks and the hypothesis-driven nature of debugging to build effective solutions.
• Your Personal Seller: Conviction sees an opportunity to empower small businesses with AI-driven tools that manage their digital presence across various platforms. Startups can develop solutions that generate optimised product listings, including descriptions, images, and videos, and A/B test them to improve conversion rates. They also suggest using AI to crawl marketplaces, identify trends, and provide recommendations for new products or untapped audiences.

Democratising Access and Ownership

• The All-Seeing Eye: Conviction highlights the need for more sophisticated tools to understand and manipulate the growing volume of video data. They see opportunities for startups to develop AI-powered video indexing and analysis solutions, particularly in the physical security monitoring industry. They suggest rethinking hardware and storage in the context of semantic video understanding and on-device models18.
• The (Smart, Social) Sims: Conviction believes that AI can create a new generation of entertainment experiences by populating virtual worlds with engaging and interesting AI characters. Startups can develop AI-driven games and platforms where users can interact with AI characters that have their own “lives,” memories, and social dynamics18. They emphasise the potential for mass personalisation, where AI generates content and experiences tailored to individual preferences.
• Accessible Government Services: Conviction sees an opportunity to use AI to simplify access to government services, making it easier for citizens to navigate complex systems and obtain necessary assistance. Startups can develop AI-powered chat interfaces that guide users through processes, answer questions, and update records, ultimately improving efficiency and reducing administrative costs.

Other Opportunities

• Knock My SOX (Compliance) Off: Conviction challenges startups to develop an AI-driven solution for Sarbanes Oxley compliance, automating tasks like controls testing, financial process analysis, and risk assessment. They believe that AI can eliminate much of the toil associated with compliance while preserving the principles of independence, monitoring, and executive responsibility.
• Save Us Money: Conviction sees an opportunity for AI agents to continuously monitor prices, aggregate demand, negotiate deals, and make purchasing decisions on behalf of consumers, similar to earlier concepts like Honey and Paribus. Startups can develop solutions that leverage web crawling, price tracking, and negotiation capabilities to help consumers save money on purchases.
• Loosening the Legacy Grip: Conviction encourages startups to challenge the conventional wisdom that legacy systems are too difficult to modernise. They specifically highlight the opportunity to modernise mainframe applications, citing the potential to unlock $100 billion in cloud spend. Startups can develop AI-powered solutions that extract business logic, translate code to modern languages, generate documentation, and map schemas, making it easier to migrate mainframe applications to the cloud.
• Many, Many, Materials: Conviction highlights the potential for AI to revolutionise materials science, citing recent research demonstrating the ability to predict stable structures and synthesise novel compounds. They see opportunities for startups to develop AI-driven solutions for material proposal and synthesis, with applications in various industries, including battery technology and energy-efficient manufacturing.
• Web Data APIs: Conviction points out the limitations of current web content APIs, which hinder the development of advanced AI applications that require access to live, reliable information. They see an opportunity for startups to build more comprehensive and flexible web data APIs that provide access to page content, parsed outlinks, edit history, and other valuable data. They suggest focusing on specific verticals and leveraging customer signals to bootstrap an index.
• Always Pick Up the Phone: Conviction identifies the missed opportunity in unanswered business calls, particularly for small businesses. They believe that advances in voice generation and LLMs are nearing the point where AI can handle many transactional calls. Startups can develop solutions that combine AI-powered voice assistants with distribution channels, customer journey design, guardrails, and workflow automation to address this need.
• Autonomous HR (and IT) Helpdesk: Conviction sees the potential for AI to dramatically improve the efficiency and employee experience of HR and IT helpdesk functions. They believe that AI-powered conversational search interfaces can replace traditional service catalogs and forms, providing employees with quick and accurate answers to their questions. Startups can develop solutions that leverage LLMs to understand and respond to natural language queries, access relevant documentation and systems of record, and automate HR and IT processes.

10 VC investment themes in 2025 according to Stripe

Stripe highlighted the biggest global trends impacting startups in 2025, focusing on evolving customer needs and the power of AI. Stripe anticipates these trends to create significant business opportunities for new and existing companies. Here are some key areas and startup opportunities:

• Generative AI and Autonomous Technology:
  • Hyper-personalisation: AI can personalise marketing campaigns and product recommendations at a scale unmatched by human teams, creating highly customised experiences for many users.
  • Autonomous Logistics: Startups can leverage AI for fleet management, self-operating warehouses, and delivery drones to optimise costs and delivery times.
  • Smart Manufacturing: AI can predict maintenance needs, automate quality control, and improve production line precision. This allows smaller manufacturers to compete with established companies.
• Decentralised Finance (DeFi) and Fintech:
  • Decentralised Financial Systems: Startups can utilise DeFi and blockchain technology to create financial systems that bypass traditional intermediaries like banks. This can be especially impactful in emerging markets where access to traditional banking is limited.
  • Blockchain-Based Lending: Smart contracts can facilitate secure lending transactions without traditional credit checks or lengthy approval processes.
  • Embedded Finance: Integrating financial services into non-financial platforms is a growing trend. Fintech startups can embed payment solutions directly into retail and service platforms. Buy Now, Pay Later (BNPL) solutions are expanding beyond traditional retail into sectors like healthcare, travel, and education.
• Logistics and Manufacturing Innovations:
  • AI-Driven Inventory Management: AI-based systems can analyse real-time data to predict demand, optimise stock levels, and reduce overproduction.
  • Supply Chain Transparency: Blockchain solutions can create traceable supply chains, building trust with shoppers and ensuring compliance, particularly in industries like food, fashion, and electronics.
• Hyper-Personalised Offerings:
  • Ecommerce and Predictive Algorithms: AI can anticipate customer desires by analysing behaviour, browsing patterns, and purchase history, enabling platforms to make personalised suggestions and create customised experiences.
  • Data-Driven Personalisation: AI can analyse individual customer data to personalise elements ranging from email marketing to product recommendations.
• Climate Technology and Sustainable Consumer Goods:
  • Green Technology: Startups focusing on clean energy solutions, such as advanced solar panels, energy storage systems, and microgrids, are poised for growth. Similarly, waste management startups using AI to optimise recycling processes present another opportunity.
  • Circular Economy: Startups developing genuinely eco-friendly alternatives are attracting attention. Vertical farming, lab-grown meat, and zero-waste packaging solutions are examples of this growing trend.
• Vertical Software-as-a-Service (SaaS) and Industry-Specific Solutions:
  • Industry-Specific Customisation: Vertical SaaS provides tailored solutions that meet the unique needs of specific sectors like legal, agriculture, and healthcare. These solutions can integrate with existing industry technology, comply with regulations, and improve workflows.
• New Applications of Web3:
  • Digital Identity Platforms: Startups can create platforms where individuals control their data and identities without reliance on centralised authorities.
  • Decentralised Social Networks: Blockchain-based social networks offer greater transparency and user control, addressing concerns associated with traditional social media platforms.
  • NFT-Based Membership Programs: NFTs can create exclusive membership programs and innovative solutions in areas like event ticketing and digital ownership verification.
  • Real Estate and Gaming: Web3 offers opportunities in real estate through tokenised property ownership and fractional ownership models. In gaming, decentralised platforms enable players to own and trade in-game assets using blockchain technology.
• Workforce Trends:
  • Hybrid Work Solutions: Startups can develop tools for managing hybrid teams, scheduling office days, and fostering collaboration across time zones.
  • Talent Marketplaces: Platforms connecting specialised freelancers and fractional workers with companies are creating new employment models.
  • Employee Wellness and Productivity Tools: Startups can address the challenges of hybrid work by developing tools focused on employee well-being and productivity.
• On-Demand Startups:
  • Dark Stores: Small, localised warehouses optimised for rapid online order fulfilment create opportunities for startups to meet consumer demand for instant gratification.
  • Instant Pay: Startups enabling real-time access to earnings for workers, particularly in the gig economy, are transforming payment systems.
  • On-Demand Healthcare: Startups providing on-demand healthcare services, such as virtual consultations and house calls, offer solutions for immediate access to care.
• Conscious Consumerism and Ethical Startups:
  • Ethical Labor Practices and Fair Trade: Startups prioritising fair wages, ethical sourcing, and transparency in supply chains are attracting a growing customer base.
  • Social Good: Companies integrating social good into their business models while maintaining profitability are demonstrating the viability of combining purpose with profit.

Requests for Gaming Startups from A16Z’s Speedrun

A16Z Games Speedrun invests in both gaming studios and technologies. Their latest request for startup proposals outlines their investment focus areas across gaming technology, platforms, and interactive experiences:

Gaming & Platform Innovation:

• Telegram game teams with expertise in distribution, monetization, and viral growth
• Fully on-chain games enabling players to fork, mod, and own the core game itself
• Crypto games developing unique digital economies
• Game studios creating genre-defining games and generational businesses
• Seeking scrappy developers with strong theses and unusual talent who push boundaries in game loops, business models, and platforms
• Looking across multiple platforms (mobile, PC/console, Discord, web3, AR/VR) and genres (roguelikes, competitive PvP, MMOs)

Interactive Technology:

• High-quality interactive experiences delivered as quickly as images or video clips
• AI-powered platforms that mimic or enhance unique real-world behavior
• Interactive platforms leveraging new technologies like AR/VR, PSP, and Meta Quest 3

Enterprise Applications:

• Founders using interactive 3D game technology for enterprise and beyond
• Interactive 3D tools for company design, visualization, and simulation needs including:
  • Product, architectural, and interior design
  • Marketing and immersive training
  • Digital twins
  • Synthetic data generation
  • Reinforcement learning in virtual worlds
• Target industries include defense, real estate, fashion, e-commerce, automotive, robotics, manufacturing, healthcare and more

The next big thing in 2025 – VC Collection

50 of technology’s top thinkers weigh in on the year ahead on Nikhil’s newsletter.

AI Agents

AI emerges as the dominant theme with predictions encompassing various aspects of its development and impact. The experts highlight the potential of AI agents, predicting their rise in different forms, including:

• Agentic apps
• AI agents capable of handling complex, multi-step tasks
• Widespread adoption of AI agents by both individuals and businesses
• AI transitioning from assistants to collaborators
• A boom in voice AI applications leading to human-like experiences
• New user interfaces that allow humans to manage work delegated to AI

These predictions highlight the need for startups to focus on developing more intuitive and sophisticated interfaces that enable seamless collaboration between humans and AI. This could include advancements in natural language processing, voice recognition, and visual interfaces that facilitate better understanding and control over AI-driven processes.

Beyond agents and interfaces, another area of high potential is the expansion of AI applications across various sectors, including:

• Consumer AI: New social apps, interactive media, and AI-driven tools for social networking and entertainment.
• AI-Generated Content: Commoditized AI content creation, particularly in video generation, will reshape how content is produced and consumed.
• Real-Time AI: The rise of real-time AI, driven by faster processing speeds, will unlock new applications requiring instant computation.
• AI for the Mainstream: AI will become more integrated into everyday products and services, seamlessly enhancing user experiences without necessarily being explicitly labeled as “AI.”

These predictions suggest a vast landscape for startups to explore. Opportunities exist in developing novel AI-powered applications for consumers, creating tools and platforms for AI content generation, and leveraging real-time AI capabilities for various use cases.

Other notable predictions and startup opportunities:

• AI-Native Hardware: The development of computers and devices specifically designed for AI, optimizing for local inference and AI interactions.
• AI-Native Organizations: The emergence of “cognicos,” companies built on an AI-native technology stack, with humans playing a more strategic role in managing and orchestrating AI-driven processes.
• Simulation as a Service: A new business model where AI models simulate decisions and conversations to optimize outcomes for individuals and businesses.
• Outcome-Based Pricing Models for AI: AI companies shifting from traditional software pricing to models based on the value and outcomes delivered.

Beyond the realm of AI

• Software Development: The rise of AI coding assistants will democratize software creation, leading to a surge in new products and applications.
• Crypto-Fintech: Applications built on stablecoins and the revival of crypto-fintech products.
• Healthcare: Continued growth in consumer healthcare, with a focus on data availability and portability, and advancements in GLP-1 drugs.
• Robotics: Advancements in general-purpose robots will transform industries reliant on physical labor.
• Defense: Increased defense spending and startup activity, driven by factors like AI advancements and geopolitical tensions.

Final thoughts

The landscape for startups in 2025 is brimming with opportunity, driven by technological advancements, evolving consumer expectations, and a global need for innovative solutions. At Remagine Ventures, we are actively seeking out those founders who can capitalise on these trends and shape the future of technology and business. It’s never to early to ask for feedback. Time to build!

• Author
• Recent Posts

Follow me

Eze Vidra

Co Founder and Managing Partner at Remagine Ventures

Eze is managing partner of Remagine Ventures, a seed fund investing in ambitious founders at the intersection of tech, entertainment, gaming and commerce with a spotlight on Israel.

I'm a former general partner at google ventures, head of Google for Entrepreneurs in Europe and founding head of Campus London, Google's first physical hub for startups.

I'm also the founder of Techbikers, a non-profit bringing together the startup ecosystem on cycling challenges in support of Room to Read. Since inception in 2012 we've built 11 schools and 50 libraries in the developing world.

Follow me

Latest posts by Eze Vidra (see all)

• Requests for startups in 2025 - January 8, 2025
• 19 years of VC Cafe - January 7, 2025
• The potential of Mini Apps to disrupt App Distribution as we know it - December 21, 2024