Talking Computers
Winter 2026 NewAI for AI Infrastructure
We deploy fleets of AI Infrastructure Engineers to collaborate over week-long time horizons, run 100s of experiments in parallel, and learn from them to optimize your training/ inference infrastructure to your workloads. We've partnered with companies to speed up their GPU kernels to 6x SOTA, and improved the latency of voice models to twice the competition.
AI Investor Summary
Talking Computers is building an AI workforce to optimize AI training and inference infrastructure, promising significant speedups and latency reductions. They are tapping into a massive and rapidly growing market at an opportune time, with early product demonstrations showing impressive results. The key challenge will be to validate these claims at scale and build a sustainable competitive advantage in a burgeoning landscape.
Key Highlights
- ● Addresses a critical and rapidly growing market need in AI infrastructure optimization.
- ● Innovative product concept leveraging AI agents for infrastructure engineering.
- ● Demonstrated impressive performance improvements in early partnerships.
Risk Factors
- ● Need to rigorously validate performance claims and demonstrate scalability across diverse workloads.
- ● Competition from established cloud providers and specialized AI optimization tools is intensifying.
- ● Building a defensible moat beyond initial performance gains will be crucial.
- ● Team needs to expand to cover sales, marketing, and broader engineering expertise.
Founders
Zayaan Mulla is the co-founder of Talking Computers, a Y Combinator startup focused on AI-powered communication solutions. His background likely includes significant experience in software development and artificial intelligence, given the company's focus. He is a graduate of the University of Waterloo.
Parsa Bahraminejad is the co-founder of Talking Computers, a Y Combinator startup focused on AI-powered communication solutions. His background likely includes significant experience in artificial intelligence, software development, and potentially product management, given his role in a tech startup. He is a graduate of the University of Toronto.
Score Breakdown
Strong technical foundation with founders from reputable universities (Waterloo CS, UofT MSc). While no prior exits or notable big tech employers are listed, their academic backgrounds suggest strong problem-solving capabilities. The small team size is typical for seed stage, but scaling will require broader expertise.
The AI infrastructure optimization market is experiencing explosive growth, driven by the massive investment in AI development and deployment. The TAM is enormous, encompassing any organization training or running AI models. Timing is excellent, as companies are actively seeking ways to reduce costs and improve performance in this critical area. Regulatory tailwinds are generally positive for AI innovation, though ethical considerations and data privacy could become factors. [Boost +0.5: Hot sector: ai]
The core concept of AI Infrastructure Engineers is technically innovative and addresses a real pain point. The claimed 6x speedup for GPU kernels and 2x latency improvement for voice models are impressive, but need further validation. The defensibility lies in the proprietary learning algorithms and the accumulated data from running experiments. UX quality is not detailed, and platform potential is high if they can generalize their optimization capabilities across various AI workloads and hardware.
Early stage with limited publicly available traction data. The partnership mentions and press coverage are positive indicators of early interest and validation, but revenue and user numbers are not disclosed. Investor interest is implied by their YC batch, but concrete commitments are not evident. This score reflects the early nature of their progress. [Boost +2: Tier-1 VC: accel]
News
Talking Computers is building Facility, a platform that serves as a workplace for AI, aiming towards fully autonomous companies by deploying AI agent workforces within enterprises.
Talking Computers has launched Facility, a platform offering companies a 24/7 parallel AI workforce composed of TC-1 AI agents that act as autonomous, self-improving coworkers.
Talking Computers is developing 'Facility,' a platform for deploying autonomous AI organizations, enabling AI agents to collaborate like a remote team.
Talking Computers is developing 'Facility, The Workplace for AI,' aiming for an AI workforce, but faces challenges in market clarity, differentiation, and product demonstration.
Talking Computers, founded in 2025, is an unfunded AI-based communication platform offering voice-based solutions, with competitors including Dialpad and LivePerson.
Talking Computers is developing 'Facility,' a platform for deploying autonomous AI organizations, aiming to create a collaborative AI workforce that operates 24/7.
Talking Computers is building AI infrastructure by deploying AI engineers to optimize training and inference infrastructure, claiming significant speed and latency improvements.
Talking Computers launched 'Facility,' a workplace for AI, featuring their TC-1 AI computer with Ambient Intelligence designed to act as a proactive, self-improving coworker.
Quick Info
- Batch
- Winter 2026
- Team Size
- 2
- Location
- Unspecified
- Founders
- 2
- Scraped
- 4/10/2026