Skip to content
Elie Bismuth

I put AI in production and generate revenue

I take an idea and ship it to production — battle-tested at Doctrine, HireSweet and Topo.io (YC 2024).

0M+
ARR generated
0
years of experience
0
SaaS shipped

Trusted by

Doctrine logoTopo logoY Combinator logoHireSweet logo

Proven results in production

Flow Counsel — AI-Powered Contract Analysis

~1M€ ARRrevenue generated400daily active users5 000analyses / day

Doctrine is France's leading legal intelligence platform, used by over 20,000 legal professionals. The company wanted to expand into the contracts space, but had no defined product direction.

Read case study

Topo — Collaborative Sales Rooms for B2B

YC W24Y Combinator1erengineer hired

Topo.io is a collaborative workspace for B2B sales teams: digital sales rooms that help sellers better engage buyers and deliver a modern buying experience.

Read case study

HireSweet — Tech Talent Matching & Recruitment Platform

300+daily active clients-50%Ops processing time

HireSweet operated two complementary products in tech recruitment: Discover, a marketplace of qualified candidates presented weekly to companies, and Reveal, a SaaS ATS to centralize and automate candidate database management.

Read case study

I don't code features. I build products.

Every technical decision I make starts with a simple question: what's the impact for the user and for the business? Choosing a framework, designing an API, integrating an AI model — everything goes through that filter. The most elegant tech in the world is worthless if it doesn't solve a real problem.

I'm constantly looking for the right balance between quality and velocity. No technical debt out of laziness, but no over-engineering out of ego either.

When I join a team, I truly integrate. I read the codebase before proposing changes. I adopt your practices before suggesting new ones. And I leave the project in a better state than I found it.

What I ship with in production

14 technologies, 3 products shipped at startups & scaleups — from API to monitoring.

Frontend / Product

  • TypeScriptEnd-to-end strict mode, from API to frontend
  • React / Next.jsProduction apps at Doctrine, Topo (YC W24), HireSweet
  • GraphQLTyped APIs with client code generation

Backend

  • Node.js / NestJSProduction microservices and modular monoliths
  • PostgreSQL / MongoDBRelational and document modeling driven by business needs
  • Queues & CachesRedis, SQS — async jobs and application caching

AI / LLM

  • OpenAI, Anthropic ClaudeProduction LLM integrations with prompt engineering
  • RAG & EmbeddingsRetrieval-augmented pipelines for domain data
  • Agents (CrewAI, LangChain, LangGraph)Multi-agent AI workflow orchestration
  • BMAD & Spec-Driven DevelopmentStructured methodology to drive AI agents end-to-end

Infrastructure / Observability

  • AWSMulti-service cloud deployments at startups
  • DatadogApplication monitoring, alerting and observability
  • SentryProduction error tracking and performance monitoring
  • Docker & KubernetesContainerization and orchestration in startup environments

Let's discuss your next AI product

30 minutes free, no commitment.