About Apex36
We ship production AI for teams who need it to actually work.
Apex36 is a Mumbai-based AI and software studio founded in 2022. We build the AI systems behind real products: RAG over millions of documents, multi-LLM routing, AI agents in production, and the full-stack SaaS infrastructure that holds it all together. We work with founders and enterprises across 5+ countries.
50+
Projects delivered
1M+
Cumulative app users
$2M+
Raised by clients we built for
98%
Client satisfaction
Our story
Apex36 started in 2022 with a simple thesis: most AI agencies sell prototypes. Real product teams need AI that survives production, meaning evaluated, observable, cost-controlled, and reliable when traffic spikes.
Since then we've shipped 50+ projects across legal tech (DecoverAI raised $2M+ seed on what we built), education (Speakology AI now runs in 600+ schools), sales intelligence (Vectro AI), and product management (Mappie AI cut grooming time by 60 to 70%). The pattern across every engagement: a founder or product team who needs AI to work, not just demo well.
We're headquartered in Mumbai, Maharashtra, India and work with teams globally. The studio is small by design, so every project gets the engineers who actually wrote the code, not a delivery manager passing messages.
Team
Yash Soni
Founder & CEO
8+ yearsFounded Apex36 Technologies in 2022 to ship production AI for product teams. With 8+ years of experience in full-stack engineering, RAG systems, and scaling SaaS infrastructure, he works with early-stage startups and enterprises across 5+ countries.
Abhishek Patel
Full Stack Engineer
1 yr 10 mosFull Stack Engineer at Apex36 who leads a team building scalable, AI-integrated web applications. Specialises in OpenAI integrations, RAG-based chatbots with real-time streaming, vector search pipelines, and production SaaS features across the full stack.
Raj Chauhan
Python Developer
1 yr 10 mosPython Developer at Apex36 focused on building robust, scalable backend systems using Python and Django. Works on LangChain-based AI pipeline integrations, e-commerce backends, and data-driven web applications with a strong emphasis on clean architecture and practical engineering.
Raj Patel
Full Stack Engineer
1 yr 5 mosFull Stack Engineer at Apex36 specialising in AI-native products: document intelligence, semantic search, data extraction pipelines, and conversational interfaces. Has shipped production systems and delivered a US client project early in his career. Brings a deliberate approach to LLM integration, knowing where models belong versus where deterministic code should take over.
Tanmay Bhole
Frontend Engineer
10 monthsFrontend Engineer at Apex36 with hands-on experience across React, Next.js, and Flutter. Previously led a mobile app development team at Google Developer Student Clubs, building cross-platform applications. Oracle-certified in Generative AI, focused on crafting web and mobile experiences that solve real-world problems.
How we work
Ship production, not prototypes
AI demos are easy. Production AI that is eval'd, observable, cost-controlled, and reliable under load is what we ship. Every engagement ends with code your team owns and can extend.
Honest before optimistic
If your AI feature isn't ready (data is too sparse, costs don't pencil out, the workflow is wrong), we tell you before you spend the budget. The free 48-hour audit exists for exactly this reason.
Small team, deep ownership
We work as an extension of your team, not a black box. Daily check-ins, transparent decisions, and direct access to the engineers writing the code.
Stay current
We publish weekly on what's actually working in production: Claude Opus, GPT-5, Gemini 3, RAG architectures, and agent systems. The blog is research notes from real engagements, not marketing.
Thinking about adding AI to your product?
Start with the free 30-minute strategy call or the 48-hour AI Integration Audit. Either way, you leave with an honest read on what to build, what to skip, and whether your data is ready.