Pursuit intelligence platform for architecture and engineering firms
Kantiv builds a pursuit intelligence platform for AEC (architecture, engineering, construction) marketing and business development teams—capturing proposals, project data, and client history to automate go/no-go decisions and proposal generation. The tech stack is heavily weighted toward LLM tooling (LangChain ecosystem, OpenAI, Mistral, Llama, RAG) and Python/Go backend services, with active projects spanning agentic proposal writing, LLM security governance, and event-driven architecture. Pain points cluster around AI deployment risk (securing LLM workloads, compliance, debugging complex AI systems) and operational efficiency (scaling SaaS, integrating ML models)—a pattern typical of early-stage AI product companies still hardening production readiness.
Kantiv (formerly Joist AI) is a SaaS platform that centralizes institutional knowledge—won proposals, client relationships, team expertise, project data—into a searchable, intelligent system for AEC firms. The platform powers pursuit workflows by automating proposal writing, surfacing relevant past work, and enabling data-driven go/no-go decisions. Founded in 2022 and headquartered in San Diego, the company operates across the United States and India. The product is built for mid-market to enterprise AEC firms where business development and marketing teams struggle to systematize knowledge capture and reuse across pursuit cycles.
Python, Go, GraphQL, FastAPI, React frontend. LLM-heavy: LangChain (LangSmith, LangGraph), RAG, OpenAI, Mistral, Llama, LlamaIndex. AWS infrastructure (Lambda, RDS, Step Functions). Testing: Playwright, TestRail, Postman. Security scanning: SAST, DAST.
Agentic proposal writing and modular agent components; LLM security governance and cloud infrastructure hardening; event-driven backend architecture; ML model integration; automated security guardrails for AI workloads in production.
Kantiv's technology stack, projects, and hiring signals are inferred from public hiring and company data — career pages, public listings, and company web presence — then clustered and de-duplicated. Figures are estimates that refresh over time. Read our full methodology →
This is not an official vendor or customer list. It is a technology-adoption signal inferred from public data, intended for B2B research.