Founding Software Engineer - AI
Posted Updated
About Dexicon:
Based in Silicon Valley but hiring remotely.
We're building the tribal knowledge context engineering layer for AI coding assistants. When one engineer solves a tricky schema issue at 2 AM, we make sure the next engineer doesn't spend 45 minutes rediscovering the same solution.
We're a VC funded, seed-stage, AI-native startup founded by operators from Amazon (One Medical), Twitch, Salesforce, and NVIDIA. Our founders have built ML infrastructure at scale and understand what it takes to ship fast without breaking things.
What You'll Work On:
- Data ingestion pipelines: Parse and structure session data from IDE logs and developer tools
- LLM-powered extraction: Build systems that extract structured knowledge from unstructured debugging sessions
- Hybrid search infrastructure: Vector + metadata retrieval engines that surface relevant context with high precision
- MCP server & API layer: Tools that AI assistants integrate with during development workflows Integration systems: Connect external data sources (GitHub, Google Docs, etc.) into our knowledge graph
Tech stack: Python, TypeScript, Vector DBs, LLMs (Claude/GPT), PostgreSQL, Next.js
Who We're Looking For:
Required:
- 2+ years of software engineering experience at a top tier software company
- Strong backend engineering fundamentals. You've built APIs, designed databases, and debugged production systems
- Experience with Python or Go in production environments
- Comfort with ambiguity. You can take vague requirements and turn them into working systems
- Strong written and verbal communication in English (async-first remote team)
Bonus points:
Experience with vector databases or semantic search systems
Familiarity with LLM APIs and prompt engineering
Understanding of developer tools, IDEs, or CI/CD workflows
Open source contributions or side projects that demonstrate craft
Why Join Now:
Ground floor at a VC-backed startup solving a real problem for engineering teams
Direct impact on product direction; your decisions shape what we build
Work with founders who've shipped production ML systems and understand technical depth
Remote-first culture built for deep work, not performative presence
The Reality Check:
We're pre-PMF. You'll be building the first version of something that doesn't exist yet.
You'll wear multiple hats and context-switch between backend, integrations, and tooling.
Fast-paced, high-autonomy environment: we expect you to own outcomes, not just tasks.
If you are not ready for a intense startup culture, this is not the role for you.