<h3><strong><img style="max-width: 100%;" src="https://wargaming.com/media/ckeditor_uploads/2025/08/26/Banner-SM-648%20x%20211-Version300ppi.png" alt="" width="648"></strong></h3><h3><strong>Job Overview</strong></h3><p>We’re building a next-generation AI stack for large-scale, session-based vehicular combat. You’ll own end-to-end AI features—from behavior technical design to integration with our server simulation, sensory systems, telemetry, and ML inference services. You’ll help us move from “default bots” to adaptive, performant, debuggable opponents and squadmates that feel human without breaking fairness or the server budget.</p><h3><strong>What will you do?</strong></h3><ul><li><strong>Design & Implement Behaviors</strong>: Ship production-grade AI using <strong>GOAP, Utility AI, and/or BTs</strong>, choosing the right tool per feature (target selection, positioning, driving, firing, evasion, cooperation).</li><li><strong>Author Sensors & World State</strong>: Build reusable <strong>sensors</strong>, perception pipelines, blackboards/knowledge graphs, and data streams (LOS/LOF, threat & opportunity scoring, terrain/cover, trafficability, dynamic obstacles).</li><li><strong>Telemetry & Tooling</strong>: Add <strong>introspection/telemetry</strong> (action traces, scores, timers, counters), record-and-replay, debug overlays/heatmaps, and scenario harnesses for regression.</li><li><strong>Performance & Quality</strong>: Profile and optimize C++ AI systems (jobs/ECS/multithreading, memory locality, cache behavior, lock contention) to hit tight frame budgets.</li><li><strong>ML Inference Hooks</strong>: Integrate with <strong>model inference services</strong>, define data contracts, and gate ML policies safely behind deterministic fallbacks.</li><li><strong>Collaboration & Mentoring</strong>: Partner with gameplay, engine, online, and data teams; review code; mentor mid/junior engineers; uphold coding standards and test discipline</li></ul><h3><strong>What are we looking for?</strong></h3><ul><li><strong>Strong C++</strong> (C++17/20) in large, real-time codebases; STL, templates, memory/perf, concurrency; solid debugging and profiling chops.</li><li><strong>Shipped features</strong> on at least one multiplayer or simulation-heavy title (AA/AAA or high-quality indie) with <strong>AI ownership</strong>.</li><li>Deep knowledge of at least one AI architecture <strong>(GOAP, Utility AI, BTs)</strong> and practical understanding of when/why to use each.</li><li>Experience building <strong>perception/sensing</strong> systems and scoring/utility functions, not just wiring nodes.</li><li>Proven ability to design <strong>debuggability</strong> in from day one (telemetry, traces, visualization, reproducible test scenarios).</li><li>Tools: <strong>Git/Perforce, CMake, Visual Studio/Clang</strong>, CI fundamentals.</li></ul><h3><strong>What additional skills will help you stand out?</strong></h3><ul><li><p><strong>Python 2.7 and 3.10</strong>+ for tooling or scripting; binding Python↔C++ and shipping safe/native bridges.</p></li><li><p>Hands-on with <strong>ML for games</strong> (data collection, labeling, offline training, online inference, A/B evaluation); familiarity with RL/imitation learning concepts.</p></li><li><p>Experience with <strong>inference runtimes</strong> (ONNX Runtime, Triton), <strong>serialization</strong> (protobuf/flatbuffers), and service plumbing (gRPC/HTTP, Kafka/Kinesis).</p></li><li><p>Navigation for vehicles at scale (terrain analysis, cost fields, flow fields, RVO/ORCA).</p></li><li><p>Prior <strong>end-to-end AI work</strong> on shipped or open-source titles (e.g., <strong>OpenTTD/Quake </strong>mods) showing full ownership from design to debugging.</p></li><li>Exposure to <strong>ECS/job systems</strong>, lock-free structures, and SIMD; Windows/Linux cross-platform builds.</li></ul><h3><strong>Tech You'll Touch</strong></h3><ul><li><strong>C++17/20</strong>, custom engines, internal ECS/job systems</li><li>Python 2.7 (tooling/scripting on one title), 3.10+ (machine learning)</li><li>Build/CI: CMake, <span data-teams="true">Teamcity</span></li><li>Observability: internal telemetry, replay tools, debug UIs/overlays</li><li>(Optional) ML: ONNX/TorchScript/TensorRT, gRPC/REST service</li></ul><h3><strong>Expected Outcomes (First 12 Months)</strong></h3><ul><li>Ship a <strong>robust AI behavior slice</strong> (e.g., target selection + driving + firing loop) with clear KPIs (engagement quality, fairness, perf budget).</li><li>Deliver a <strong>sensor & scoring framework</strong> that other engineers/designers can extend safely.</li><li>Stand up <strong>AI telemetry & replay</strong> that shortens bug reproduction cycles from days to hours.</li><li>Integrate and harden <strong>one ML-backed component</strong> behind toggles and fallbacks.</li></ul><h3><strong>What to Include with Your Application</strong></h3><ul><li>Short note on <strong>an AI system you owned</strong>: problem, architecture choice, trade-offs, and how you measured success.</li><li>Links to <strong>demos, videos, or repos</strong> (including open-source or modding AI work).</li><li>(Optional) A brief example of a <strong>telemetry view or replay harness</strong> you built.</li></ul><h3><strong>Work mode</strong></h3><ul><li>Remote (from anywhere within the selected country/EU).</li></ul><h3><strong>Benefits</strong></h3><p>Benefits and perks are tailored to the local market and culture. Our benefits in Berlin include:</p><ul><li>5 weeks of Annual Leave per calendar year</li><li>Additional paid time off </li><li>Career development and education opportunities within the company</li><li>Home internet reimbursement</li><li>Gym membership through Urban Sports Club - Package M</li><li>Mental well-being program (iFeel)</li><li>Monthly public transportation ticket</li><li>Language classes</li><li>Company pension scheme with a company contribution up to 150 EUR per month</li><li>Personal Gaming Account </li><li>Coffee, fruits, and snacks in the office</li><li>Company events</li><li>Seniority Awards </li><li>Referral program - You can recommend the best talents to the company and receive a reward </li></ul><h5>Please submit your CV in English to ensure smooth processing and review.</h5><div class="content-conclusion"><h3>About Wargaming </h3><p>Wargaming is an award-winning online game developer and publisher headquartered in Nicosia, Cyprus. Operating since 1998, Wargaming has become one of the leaders in the gaming industry with 15 offices worldwide, including studios in Chicago, Prague, Shanghai, Tokyo, and Vilnius. Our diverse and multicultural team works together to deliver a top-class experience to millions of players who enjoy Wargaming's titles across all major gaming platforms. Our flagship products include free-to-play hits World of Tanks, World of Warships and World of Tanks Blitz.</p><p>Please see <a href="http://legal.eu.wargaming.net/en/candidate-privacy-policy/">Wargaming Candidate Privacy Policy</a> for details on how Wargaming uses your personal data.</p></div>