Location
Kraków (5 days in office) or US (remote)
CodeStrapOpen position
Senior Engineer — Knowledge & Retrieval
Build the retrieval layer that gives Larry AI coding agents the right codebase context automatically — so engineers never have to manually assemble what the repository already knows.
Team
Knowledge Team
Level
Expert
Type
Full-time
Category
- Search
- AI
- TypeScript
Must have
- TypeScript
- Information Retrieval
- Graphs
- Databases
- Static Analysis
- LLM
- Monorepos
- English (B2+)
Nice to have
- Embeddings
- Python
- Nx
Senior Engineer (Knowledge & Retrieval)
Close the gap between what a codebase already knows and what an AI coding agent actually receives when an engineer asks for help.
🧠 Role overview
At CodeStrap, we do not just build software; we engineer competitive advantages. We are building Larry AI — a proprietary, near-deterministic AI coding platform designed for broad developer adoption. We are playing in the same league as Cursor and Anthropic.
We are looking for a Senior Engineer to join our Knowledge Team. This is not a role for someone who only wants to tune prompts or bolt on a vector database. This is a role for an expert individual contributor who takes Extreme Ownership over a hard problem: when an engineer asks Larry to build a component, generate a feature, or change existing code, the agent must automatically retrieve the right context — app conventions, connected components, dependencies, patterns, and architectural boundaries — without anyone manually copying it into a prompt.
We already LLMify our repositories into structured, mostly deterministic knowledge graphs. You will take ownership of LLMify and evolve it — making that knowledge searchable at scale and wired into agents at runtime, from single-package retrieval to workspace-wide index, from lexical routing to hybrid search where it earns its keep.
You will collaborate closely with Core, Platform, and Infra teams — and contribute outside Knowledge when the problem demands it.
If you operate with absolute certainty, demand uncompromising quality, and want to solve one of the defining problems in AI-assisted development, this is where you step up.
✅ Requirements
💼 Experience
- 8+ years of hard-core software engineering experience
- Proven track record as a Senior or Expert-level individual contributor
- Confidence owning complex systems from design through production
🏗️ Problem-solving depth
- Strong foundation in information retrieval — indexing, ranking, query parsing, and context budgeting
- Comfort working with graphs and databases — model dependency and knowledge relationships as graphs, persist indices at scale, and retrieve results fast under real query load
- Experience extracting structure from code — exports, dependencies, module boundaries, import relationships
- Hands-on experience assembling LLM context from retrieved sources, not dumping whole repositories into prompts
- Comfort reasoning about large monorepos — what to index, what to expand, and what to leave out
- Ability to keep deterministic foundations trustworthy while adding smarter retrieval on top
🔥 Extreme ownership
- You do not wait for tasks to be assigned
- You see a problem, design a solution, align resources, and deliver with certainty
- You naturally take accountability for outcomes, not only implementation details
🤝 Mentoring
- Ability to elevate the team through high standards and strong technical guidance
- Confidence collaborating hands-on in the office — pair programming, design discussions, and raising the bar through direct technical exchange
- Skill in helping engineers grow into independent problem solvers
📈 Business DNA
- Understanding that code serves business goals and ROI, not just technical elegance
- Ability to create durable operating leverage through technical decisions
- Strong instinct for connecting architecture choices with strategic value
🌍 What you can expect
- Frontier product: Building the knowledge layer behind Larry AI and x-reason — the retrieval engine that makes near-deterministic agents actually useful on real codebases
- Elite environment: Working alongside C-level executives, visionaries, and core architects with no corporate theater
🧩 Responsibilities
- Agent context retrieval: Design and build the pipeline that turns an engineer's natural-language request into a ranked, budgeted context pack — files, symbols, patterns, and dependency expansions — ready for coding agents to consume.
- Repository indexing: Evolve deterministic codebase indexing from structured documentation and static analysis into fast, incremental search indices that stay in sync as repositories change.
- Search and ranking: Implement retrieval modes that route queries to the right knowledge zones, score relevance, and expand along dependency graphs without losing precision.
- Workspace-scale search: Scale retrieval from single packages to entire monorepos — cross-boundary queries, merged indices, and performance that holds up on large codebases.
- Cross-team integration: Wire retrieval into agent runtimes and platform services; collaborate with Core, Platform, and Infra to ship retrieval as a first-class product capability.
🧪 Recruitment process
- 1.5h pair programming with the CodeStrap core team
- 1h meeting with the Principal Engineer
- Optional additional 1h if we want to explore fit for another team or skillset
- Decision and offer
🏢 Perks in the office
- ☕ Free coffee
- 🚲 Bike parking
- 🚿 Shower
- 🥤 Free beverages
- 🍿 Free snacks
- 🏙️ Modern office
- 👕 No dress code
- 🎮 Playroom
🎁 Benefits
- 🎤 Conference budget
- 🕒 Flexible working hours
Application form
Interested in Senior Engineer — Knowledge & Retrieval?
Share your background and contact details. The form validates on the client, the API validates the same payload, and Google reCAPTCHA helps block spam.