{"id":3727385,"url":"https://alion.io/job/stepfun-pre-trainjichumoxingtpm","title":"Pre train基础模型TPM","company":{"id":42724,"name":"StepFun","domain":"stepfun.com","url":"https://alion.io/company/stepfun-com","size_band":"51-200","is_staffing_agency":false,"employer_type":"direct","is_intermediary":false,"listed_via":null,"ats_vendor":"Moka","truth_index":{"grade":"A","score":93,"open_postings":210,"ghost_share":0,"stale_share":0.095,"repost_share":0,"time_to_fill_p50_days":111,"computed_at":"2026-11-07T05:47:08Z"}},"role":"Product","role_family":"Product","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":["Beijing, China"],"countries":["CN"],"hiring_countries":[],"hiring_countries_total":0,"salary":null,"salary_estimate":null,"experience_years_min":3,"visa_sponsorship":false,"relocation_package":false,"has_equity":false,"technologies":[{"name":"DeepSpeed","optional":false},{"name":"Megatron-LM","optional":false},{"name":"MLFlow","optional":false},{"name":"PyTorch","optional":false},{"name":"Tokenization","optional":false},{"name":"Weights & Biases","optional":false}],"status":"live","first_seen_at":"2026-07-10T00:00:00Z","employer_posted_date":"2026-11-04","last_verified_at":"2026-11-08T02:34:54Z","board_verified":true,"closed_at":null,"days_open":121,"trust":{"level":"ok","repost_count":null,"flags":[],"days_open":121},"description":"岗位职责\n负责基础模型预训练项目的端到端交付：从立项、数据准备、算力排期、训练启动到模型验收与复盘，制定里程碑与排期，把控进度、质量与风险; \n拉通跨团队协作：协调预训练数据工程、算法、训练框架 / Infra、算力资源、评测、安全合规等多方，梳理依赖关系、识别关键路径、推动阻塞问题解决; \n建立并沉淀预训练项目的流程机制，包括数据准入与配比评审、训练变更管理、实验与 Checkpoint 记录规范、故障响应与复盘（Postmortem）、模型发版流程; \n跟踪大规模训练任务（千卡 / 万卡级）的运行状态，关注算力利用率、训练吞吐、loss 异常与中断等风险信号，组织故障止损与恢复预案，维护项目风险台账; \n搭建项目度量体系与看板：数据规模与质量、MFU / 训练成本、实验迭代效率、交付准时率等，用数据支撑决策与资源诉求; \n负责向上与跨部门沟通，定期同步项目进展、风险与取舍，组织技术评审与阶段复盘，沉淀可复用的方法论与文档。\n任职要求\n本科及以上，计算机、软件工程、人工智能等相关专业，或具备同等技术背景; 3 年以上技术项目管理 / 研发管理经验，主导过大型 AI、分布式系统或基础设施类项目; \n理解大模型预训练全流程：数据采集与清洗、Tokenizer 与数据配比、分布式训练（TP / PP / DP / ZeRO）、Checkpoint 管理、评测与发布，能与算法和 Infra 团队进行实质性的技术对话; \n具备「技术判断力」：能读懂训练日志、实验配置与评测报告，识别风险信号，不需要亲手训模型，但要能判断项目是否跑偏; \n有大规模算力 / GPU 集群项目管理经验，理解资源排期、配额、成本与效率之间的约束与取舍; \n强跨团队推动力与沟通协调能力，能在无直接汇报关系的情况下驱动交付; 结构化思维清晰，能产出高质量技术文档与管理层汇报; \n熟练使用项目管理工具与数据看板，能适应高频迭代、不确定性强的研发节奏。\n加分项\n完整参与过千亿 / 万亿参数级模型预训练项目，或大规模预训练数据工程项目; \n有算法、MLE、Infra / SRE 背景后转型 TPM，熟悉 PyTorch / Megatron / DeepSpeed、Ray / Spark、WandB / MLflow 等工具栈; \n有从 0 到 1 搭建研发流程、实验管理规范或项目度量体系的经验; \n有大规模集群稳定性治理、训练中断恢复、算力成本优化的项目经历;","description_format":"text","description_chars":1027,"description_truncated":false,"requirements":{"experience_years_min":3,"management_years_min":null,"team_size_min":null,"manages_managers":false,"education":null,"security_clearance":false,"languages":[{"language":"Chinese","level":"Advanced (C1)","optional":false}]},"benefits":[],"hiring_locations":[],"hiring_excludes":[],"relocation_offered":false,"industries":[],"lifecycle":[{"event":"open","at":"2026-11-04T06:15:01Z"}],"visa":[],"liveness":{"score":35,"band":"fade","label":"Fading","p_open":1,"p_active":0.799,"p_room":0.44,"age_days":120,"expected_fill_days":111,"reasons":["conf:4","velocity","win:tail","crowd:brand"],"computed_at":"2026-11-07T05:47:08Z"},"pay":null,"html_url":"https://alion.io/job/stepfun-pre-trainjichumoxingtpm","json_url":"https://alion.io/job/stepfun-pre-trainjichumoxingtpm.json","meta":{"generated_at":"2026-11-08T05:19:13Z","cache_seconds":300,"methodology":"https://alion.io/methodology","terms":"https://alion.io/terms","contact":"https://alion.io/contact","api":"https://alion.io/developers","about":"Alion is a live layer of people, companies and AI agents: who they are, whether they are real and active right now, what they do and how to work with them, readable by people and by agents and paid per call.","catalog":"https://alion.io/catalog.json","usage":{"tier":"crawler_verified","counted_by":"address","units_charged":1,"used_today":11550,"day_limit":null,"remaining_today":null,"minute_limit":300,"resets_at":"2026-11-09T00:00:00Z"}},"offers":[{"id":"company.slices","title":"One company in depth, by slice","status":"live","price":{"credits":0.02,"usd":0.002,"plus_per_slice":{"credits":0.05,"usd":0.005}},"unit":"per company, plus each slice with data","note":"the employer in depth","call":{"mcp_tool":"get_company","arguments":{"id":42724},"rest":"https://alion.io/mcp/rest/get_company?id=42724"},"human":"https://alion.io/catalog?offer=company.slices&for=job%2Fstepfun-pre-trainjichumoxingtpm"},{"id":"market.stats","title":"A market slice: pay, demand and time to fill","status":"live","price":{"credits":1,"usd":0.1},"unit":"per slice","note":"pay, demand and time to fill for this role and place","call":{"mcp_tool":"market_stats"},"human":"https://alion.io/catalog?offer=market.stats&for=job%2Fstepfun-pre-trainjichumoxingtpm"},{"id":"job.search","title":"Open jobs by role, technology, place, pay and visa","status":"live","price":{"credits":0.02,"usd":0.002},"unit":"per posting in a list","note":"similar open postings","call":{"mcp_tool":"search_jobs"},"human":"https://alion.io/catalog?offer=job.search&for=job%2Fstepfun-pre-trainjichumoxingtpm"},{"id":"company.verify","title":"Is this company real and active right now","status":"pilot","price":null,"unit":"per company","request":{"url":"https://alion.io/catalog/request","method":"POST","body":"{\"offer\": \"company.verify\", \"for\": \"job/stepfun-pre-trainjichumoxingtpm\", \"note\": \"what you need it for\"}"},"human":"https://alion.io/catalog?offer=company.verify&for=job%2Fstepfun-pre-trainjichumoxingtpm"}],"agent":{"mcp":"https://alion.io/mcp/jobs","mcp_open":"https://alion.io/mcp/open","next":[{"tool":"get_job","arguments":{"id":3727385}},{"tool":"search_jobs","arguments":{"company":"stepfun.com","detail":"compact"}},{"tool":"match_people","arguments":{"job_id":3727385},"note":"people open to offers who fit this vacancy"}],"note":"The same data through MCP tools: https://alion.io/mcp/open needs no sign-in (compact searches, postings, people); https://alion.io/mcp with an Alion account has everything. Docs: https://alion.io/developers"}}