- Position: Senior Engineer (Data & Context Architecture)
- Type of contract: B2B
- Working model: on-site only
Role overview:
Our client does not just build software; they engineer competitive advantages. They are building a proprietary, near-deterministic AI-powered development platform designed for broad developer adoption, operating at the absolute frontier of technology alongside the world’s leading AI labs.
They are looking for a Senior Engineer to join their core data and context structuring team. This is a role for an expert individual contributor who takes Extreme Ownership over a foundational challenge: bridging the gap between what a codebase contains and what an AI development agent receives when an engineer requests assistance. When an engineer interacts with the system to build components, generate features, or modify code, the agent must automatically retrieve the perfect context—including app conventions, connected components, dependencies, architectural boundaries, and design patterns—without anyone manually copying data into a prompt. They transform complex software repositories into structured, mostly deterministic knowledge graphs. In this role, you will take full ownership of this indexing and conversion layer, evolving it to make codebase knowledge searchable at scale and fluidly wired into agents at runtime. Your work will span from single-package retrieval pipelines to workspace-wide cross-boundary indexing, leveraging everything from lexical routing to sophisticated hybrid search engines.
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 initial design through production deployment.
Technical Depth & Skills
- Strong foundation in information retrieval concepts—including indexing strategies, ranking algorithms, query parsing, and precise context budgeting.
- Comfort working with graphs and databases to model complex software dependencies and codebase relationships, persisting indices at scale, and retrieving results fast under real-time production query loads.
- Deep experience extracting structure from code, including exports, module boundaries, import relationships, and abstract syntax representations.
- Hands-on expertise assembling optimized context blocks from retrieved sources rather than dumping raw repositories into large prompts.
- Ability to reason strategically about large monorepos, determining exactly what to index, what to expand dynamically, and what to leave out.
- Skill in keeping deterministic architectural foundations completely trustworthy while building intelligent retrieval pipelines on top.
Extreme ownership
- You do not wait for tasks to be assigned; you proactively identify gaps and structural challenges.
- You see a complex engineering problem, design a clean solution, align the necessary resources, and deliver with absolute certainty.
- You naturally take accountability for final outcomes and user experiences, not just isolated implementation details.
Mentoring & Collaboration
- Ability to elevate the entire engineering team through high technical standards and clear architectural guidance.
- Confidence collaborating hands-on in a shared office environment through pair programming, interactive design discussions, and direct exchange.
- Skill in helping fellow engineers grow into entirely independent problem solvers.
Business DNA
- Deep understanding that code serves core business objectives and strategic ROI, not just technical elegance.
- Ability to create durable operating leverage through sound technical choices and future-proof design.
- Strong instinct for connecting low-level system architecture choices directly with company value.
What you can expect
- Frontier Product: Building the core system data layer and retrieval engine that makes near-deterministic AI agents highly effective and deployable across massive production codebases.
- Elite Environment: Working alongside C-level executives, industry visionaries, and core platform architects with no corporate theater or bureaucratic overhead.
Responsibilities:
- Agent Context Retrieval: Design and build the pipeline that transforms an engineer’s natural-language request into a perfectly ranked, budgeted context pack consisting of symbols, files, patterns, and dependency expansions.
- Repository Indexing: Evolve deterministic codebase indexing from static analysis and structured documentation into ultra-fast, incremental search indices that stay synced as codebases change in real time.
- Search and Ranking: Implement advanced retrieval modes that route queries to optimal zones, score relevance accurately, and expand along dependency graphs without sacrificing precision.
- Workspace-Scale Search: Scale retrieval mechanisms from single packages to massive enterprise monorepos, ensuring cross-boundary queries and merged indices maintain high performance.
- Cross-Team Integration: Wire retrieval pipelines into agent runtimes and core platform services, collaborating closely with infrastructure teams to ship retrieval as a first-class product feature.
Recruitment process:
- 1.5-hour pair programming session with our client’s core engineering team.
- 1-hour technical interview with the Principal Engineer.
- Optional additional 1-hour conversation to explore deeper fit or alternative skillsets.
- Final decision and offer.
Perks in the office:
- Premium coffee selection
- Secure bike parking
- On-site shower facilities
- Free selection of beverages
- Complimentary snacks
- Modern, high-end office space
- No dress code policy
- Dedicated playroom and relaxation area
Benefits:
- Dedicated conference and professional development budget
- Flexible working hours
