{"id":1289961,"url":"https://alion.io/job/uforce-staff-ai-engineer","title":"Staff AI Engineer","company":{"id":1922281,"name":"UFORCE","domain":"uforce.com","url":"https://alion.io/company/uforce-com","size_band":"11-50","is_staffing_agency":false,"employer_type":"direct","is_intermediary":false,"listed_via":null,"ats_vendor":"Ashby","truth_index":null},"role":"AI/ML","role_family":"AI/ML","seniority":"staff","employment_type":"full_time","work_mode":"on_site","remote_scope":null,"remote_scope_basis":null,"remote_working_hours":null,"hiring_geo_confidence":"structured","locations":["London, United Kingdom"],"countries":["GB"],"hiring_countries":[],"hiring_countries_total":0,"salary":null,"salary_estimate":{"min_usd":140000,"max_usd":273000,"period":"year","method":"role_seniority_country_remote_cell","sample_n":14},"experience_years_min":5,"visa_sponsorship":false,"relocation_package":false,"has_equity":false,"technologies":[{"name":"Computer Vision","optional":false},{"name":"Grounding DINO","optional":false},{"name":"Human-in-the-Loop","optional":false},{"name":"LLM","optional":false},{"name":"Python","optional":false},{"name":"PyTorch","optional":false},{"name":"Synthetic Data","optional":false},{"name":"Context Engineering","optional":true},{"name":"Dagster","optional":true},{"name":"Federated Learning","optional":true},{"name":"Flyte","optional":true},{"name":"Kubeflow","optional":true},{"name":"Kubernetes","optional":true},{"name":"MLFlow","optional":true},{"name":"Sensor Fusion","optional":true},{"name":"Sim-to-Real","optional":true},{"name":"Structured Outputs","optional":true},{"name":"Weights & Biases","optional":true}],"status":"live","first_seen_at":"2026-08-27T10:39:29Z","employer_posted_date":"2026-08-27","last_verified_at":"2026-09-26T20:55:34Z","board_verified":true,"closed_at":null,"days_open":30,"trust":{"level":"ok","repost_count":null,"flags":[],"days_open":30},"description":"UFORCE exists to make aggression unaffordable.\nFounded in Ukraine and headquartered in London, UFORCE is a defence-technology company with operations across Europe, the United States and Asia.\nUFORCE builds uncrewed vessels, ground robotics and aircraft, autonomy software, and the command-and-control layer that operates them together. Each platform is deployable on its own, and stronger as part of the system. Its product lines include the MAGURA family of uncrewed surface vessels, the NEMESIS family of strike platforms, the LIUT family of ground robotic platforms, and counter-UAS systems.\nAbout the role\nThe Staff Engineer, AI / ML - Self-Serve Toolchain will build the end-to-end system that lets customers adapt UFORCE models on their own private data without exposing that data to us.\nThis role spans data processing, foundation-model-assisted labeling, human-in-the-loop QA, active learning, training, evaluation, and model promotion. You will turn an in-flight principal-led capability into a repeatable toolchain that non-expert customers can run safely on-site.\nWhat you’ll do\nOwn the ML adaptation pipeline from raw customer data to trained, evaluated, deployable models.\n\nBuild foundation-model-assisted labeling workflows using tools such as SAM-2, Grounding DINO, open-vocabulary models, LLM steering, and human review.\n\nDesign self-serve operator workflows using yes/no/maybe feedback and natural-language corrections.\n\nCreate versioned datasets with lineage, data cards, label provenance, class distributions, and known gaps.\n\nBuild QA methods that catch systematic pseudo-label errors, missing annotations, and long-tail data gaps.\n\nDevelop active-learning loops that prioritize the highest-value frames for limited operator review.\n\nBuild reproducible train/eval pipelines with experiment tracking, model packaging, and promotion gates.\n\nDesign evaluation around held-out anchor sets, leakage prevention, baselines, slice metrics, and automated approve/reject decisions.\n\nPackage the toolchain for on-prem, air-gapped, regulated, or customer-held environments.\n\nClose the field-failure loop by feeding live failures back into the next tune cycle.\n\nLead and grow a small specialist team around curation, auto-labeling, deployment, and onboarding.\n\nWhat success looks like\nCustomers can run a full adaptation cycle without engineer intervention.\n\nCustomer data stays inside the customer boundary.\n\nThe system produces trusted datasets with clear provenance and quality signals.\n\nPseudo-label quality is measured and systematic errors are caught early.\n\nThe promotion gate can approve or reject models based on evidence, not intuition.\n\nEvaluation is protected by anchor sets, leakage controls, slice metrics, and baseline comparisons.\n\nField failures become reproducible inputs to the next training cycle.\n\nSynthetic data is used only when it proves value against real held-out data.\n\nThe toolchain becomes a repeatable capability supported by a small, ramped team.\n\nRequired Qualifications\n5+ years building production ML, AI, or computer-vision systems.\n\nStrong Python and PyTorch.\n\nExperience owning ML data pipelines, training pipelines, or evaluation infrastructure.\n\nDeep CV data experience: detection, segmentation, annotation taxonomies, dataset curation, and data quality.\n\nHands-on experience with model-in-the-loop or foundation-model-assisted labeling.\n\nFamiliarity with tools such as SAM-2, Grounding DINO, FiftyOne, and annotation platforms.\n\nStrong understanding of evaluation: held-out sets, leakage prevention, baselines, slice metrics, and promotion gates.\n\nExperience with QA sampling, label-noise analysis, missing annotations, IAA, or alternatives when only one operator is available.\n\nExperience with active learning or other methods for prioritizing labeling effort.\n\nAbility to reason about dataset economics: quality vs. quantity, long-tail coverage, and cost-per-useful-example.\n\nExperience with dataset versioning, lineage, experiment tracking, model registries, or data cards.\n\nStrong ownership, communication, and systems thinking.\n\nExperience leading engineers as a tech lead, staff engineer, or small-team manager.\n\nNice to have\nSynthetic data, sim2real, or domain randomization experience.\n\nEO / IR / LWIR, remote sensing, maritime imagery, or small-object detection experience.\n\nPrivacy-preserving ML, federated learning, on-prem, or air-gapped deployment experience.\n\nExperience building self-serve ML platforms or tools for non-expert users.\n\nExperience with lakeFS, DVC, MLflow, Weights & Biases, Kubernetes, Kubeflow, Flyte, Dagster, Airflow, or Argo.\n\nLLM application, context engineering, structured output, or LLM evaluation experience.\n\nExposure to radar, AIS, EO/IR fusion, tracking, sensor fusion, robotics, autonomy, UxV, defence tech, C2/C4ISR, or tactical systems.\n\nThe nature of combat has changed.\nTomorrow’s battlefield success depends on autonomy, speed, and adaptability.\nAnd UFORCE is ready.\nAre You?","description_format":"text","description_chars":4972,"description_truncated":false,"requirements":{"experience_years_min":5,"management_years_min":null,"team_size_min":null,"manages_managers":false,"education":{"level":"phd","optional":false},"security_clearance":false,"languages":[]},"benefits":[],"hiring_locations":[{"name":"United Kingdom","iso":"GB","kind":"country"}],"hiring_excludes":[],"relocation_offered":false,"industries":["Cybersecurity","Military","Penetration Testing"],"lifecycle":[{"event":"open","at":"2026-09-26T07:41:39Z"}],"liveness":{"score":35,"band":"fade","label":"Fading","p_open":1,"p_active":0.473,"p_room":0.75,"age_days":30,"expected_fill_days":31,"reasons":["conf:3","win:late"],"computed_at":"2026-09-27T00:01:52Z"},"pay":null,"html_url":"https://alion.io/job/uforce-staff-ai-engineer","json_url":"https://alion.io/job/uforce-staff-ai-engineer.json","meta":{"generated_at":"2026-09-27T00:01:52Z","cache_seconds":300,"methodology":"https://alion.io/methodology","terms":"https://alion.io/terms","contact":"https://alion.io/contact","api":"https://alion.io/developers","usage":{"tier":"crawler","counted_by":"address","units_charged":1,"used_today":7,"day_limit":5000,"remaining_today":4993,"minute_limit":60,"resets_at":"2026-09-28T00:00:00Z"}}}