{"id":1222512,"url":"https://alion.io/job/dunia-ml-engineer-agents-reasoning","title":"ML Engineer, Agents & Reasoning","company":{"id":2222519,"name":"Dunia","domain":"dunia.ai","url":"https://alion.io/company/dunia-ai","size_band":"11-50","is_staffing_agency":false,"employer_type":"direct","is_intermediary":false,"listed_via":null,"ats_vendor":"Personio","truth_index":{"grade":"D","score":40,"open_postings":10,"ghost_share":1,"stale_share":0,"repost_share":0,"time_to_fill_p50_days":null,"computed_at":"2026-09-27T05:45:00Z"}},"role":"AI/ML","role_family":"AI/ML","seniority":"middle","employment_type":"full_time","work_mode":"on_site","remote_scope":null,"remote_scope_basis":null,"remote_working_hours":null,"hiring_geo_confidence":"structured","locations":["Berlin, Germany"],"countries":["DE"],"hiring_countries":[],"hiring_countries_total":0,"salary":null,"salary_estimate":{"min_usd":71000,"max_usd":189000,"period":"year","method":"global_role_cell_scaled_by_country","sample_n":456},"experience_years_min":4,"visa_sponsorship":false,"relocation_package":false,"has_equity":false,"technologies":[{"name":"AI Agents","optional":false},{"name":"EU AI Act","optional":false},{"name":"JAX","optional":false},{"name":"PyTorch","optional":false}],"status":"live","first_seen_at":"2026-02-06T13:55:16Z","employer_posted_date":"2026-02-06","last_verified_at":"2026-09-27T10:12:08Z","board_verified":true,"closed_at":null,"days_open":233,"trust":{"level":"ghost","repost_count":0,"flags":["stale","company_stale"],"days_open":232},"description":"Your mission\nBuild agentic AI systems that reason, plan, and act inside real materials discovery workflows Most agent systems live in clean environments: browsers, codebases, or synthetic benchmarks. At Dunia, agents must reason about messy reality: experiments that fail, data that contradicts itself, and physical systems that don’t reset cleanly.\nAs ML Engineer, Agents & Reasoning, you build the systems that make AI act responsibly inside that reality. You design agents that decide what to do next, use tools intelligently, recover from failure, and know when they don’t know.\nYour work sits at the boundary between cognition and control.\nYour tasks will include:\nBuild agentic decision-making systems for discovery \nDesign and implement agentic systems that plan, reason, and act across materials discovery workflows\nDevelop agents that operate over experiments, simulations, and scientific datasets, selecting next actions under uncertainty\nDefine how autonomy is scoped, when humans stay in the loop, and how decisions are escalated\nGround reasoning in scientific and physical reality \nImplement planning, control logic, and uncertainty-aware decision-making tailored to physical systems\nEncode operational, experimental, and safety constraints directly into agent behavior\nDefine stopping criteria, fallback strategies, and recovery mechanisms to prevent brittle behavior\nTurn models into action \nCollaborate closely with AI researchers to embed predictive models into agent workflows\nWork with lab, automation, and software teams to connect agents to real experimental and simulation systems\nEnsure agent outputs translate into executable actions, not just recommendations\nMeasure what matters \nBuild evaluation frameworks that assess decision quality, learning efficiency, and system behavior, not just model accuracy\nAnalyze failure cases and iterate on system design based on real-world outcomes\nHelp define what “good decisions” mean in scientific discovery contexts\nShip reliable, production-grade systems \nTranslate research concepts into robust, maintainable ML systems\nInstrument agents with logging, monitoring, and diagnostics for observability and debugging\nTake ownership of systems from prototype through deployment and operation\nYour profile\n4-8 years of experience building ML-driven or algorithmic decision-making systems in production or applied research settings\nStrong background in scientific or structured data modeling, rather than language-first systems\nExperience with planning, control, optimization, probabilistic reasoning, or decision-making under uncertainty\nProficiency in modern ML frameworks (e.g.PyTorch, JAX) and strong general software engineering skills\nComfortable owning systems end-to-end, from prototype to reliable operation\nAble to reason clearly about system behavior in complex, partially observable environments\nTechnically curious, with interest in physical systems, experiments, and real-world constraints\nClear communicator who can work effectively across AI, engineering, and scientific teams\nEnglish fluency;additional language desirable","description_format":"text","description_chars":3098,"description_truncated":false,"requirements":{"experience_years_min":4,"management_years_min":null,"team_size_min":null,"manages_managers":false,"education":null,"security_clearance":false,"languages":[{"language":"English","level":"Advanced (C1)","optional":false}]},"benefits":[],"hiring_locations":[],"hiring_excludes":[],"relocation_offered":false,"industries":[],"lifecycle":[{"event":"open","at":"2026-09-25T12:38:24Z"}],"liveness":{"score":3,"band":"cold","label":"Long shot","p_open":1,"p_active":0.122,"p_room":0.28,"age_days":232,"expected_fill_days":21,"reasons":["conf:10","stale_co","ghost","win:tail","crowd:"],"computed_at":"2026-09-27T05:45:00Z"},"pay":null,"html_url":"https://alion.io/job/dunia-ml-engineer-agents-reasoning","json_url":"https://alion.io/job/dunia-ml-engineer-agents-reasoning.json","meta":{"generated_at":"2026-09-28T00:53:41Z","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":440,"day_limit":5000,"remaining_today":4560,"minute_limit":60,"resets_at":"2026-09-29T00:00:00Z"}}}