{"id":4658812,"url":"https://alion.io/job/gigaforce-software-staff-ai-engineer-open-source-genai","title":"Staff AI Engineer - Open-Source GenAI","company":{"id":3800764,"name":"Gigaforce Software","domain":"gigaforce.io","url":"https://alion.io/company/gigaforce-software","size_band":null,"is_staffing_agency":false,"employer_type":"direct","is_intermediary":false,"listed_via":null,"ats_vendor":null,"truth_index":null},"role":"AI/ML","role_family":"AI/ML","seniority":"staff","employment_type":null,"work_mode":"hybrid","remote_scope":null,"remote_scope_basis":null,"remote_working_hours":null,"hiring_geo_confidence":"structured","locations":["Noida, India"],"countries":["IN"],"hiring_countries":[],"hiring_countries_total":0,"salary":{"min":1500000,"max":3000000,"currency":"INR","period":"year","gross":true,"usd_annual":32157},"salary_estimate":null,"experience_years_min":6,"visa_sponsorship":false,"relocation_package":false,"has_equity":false,"technologies":[{"name":"AI Agents","optional":false},{"name":"Angular","optional":false},{"name":"Anthropic","optional":false},{"name":"Chroma","optional":false},{"name":"Embeddings","optional":false},{"name":"FAISS","optional":false},{"name":"Fine-tuning","optional":false},{"name":"Gemini","optional":false},{"name":"Gemma","optional":false},{"name":"Hugging Face","optional":false},{"name":"Human-in-the-Loop","optional":false},{"name":"Java","optional":false},{"name":"Knowledge Distillation","optional":false},{"name":"Kotlin","optional":false},{"name":"Llama","optional":false},{"name":"LLM","optional":false},{"name":"LLM Evaluation","optional":false},{"name":"LLM Guardrails","optional":false},{"name":"LoRA","optional":false},{"name":"Machine Learning","optional":false},{"name":"Milvus","optional":false},{"name":"Mistral","optional":false},{"name":"Model Context Protocol","optional":false},{"name":"Model Distillation","optional":false},{"name":"Node JS","optional":false},{"name":"OCR","optional":false},{"name":"OpenAI","optional":false},{"name":"PEFT","optional":false},{"name":"pgvector","optional":false},{"name":"Python","optional":false},{"name":"PyTorch","optional":false},{"name":"Qdrant","optional":false},{"name":"QLoRA","optional":false},{"name":"Quantization","optional":false},{"name":"Qwen","optional":false},{"name":"RAG","optional":false},{"name":"SGLang","optional":false},{"name":"TGI","optional":false},{"name":"Transformers","optional":false},{"name":"TRL","optional":false},{"name":"vLLM","optional":false},{"name":"Weaviate","optional":false},{"name":"Docker","optional":true},{"name":"GitHub","optional":true},{"name":"JavaScript","optional":true},{"name":"Kubernetes","optional":true},{"name":"LangChain","optional":true},{"name":"LangGraph","optional":true},{"name":"LlamaIndex","optional":true},{"name":"MLFlow","optional":true},{"name":"Multimodal AI","optional":true},{"name":"NLP","optional":true},{"name":"PostgreSQL","optional":true},{"name":"Synthetic Data","optional":true},{"name":"TypeScript","optional":true}],"status":"live","first_seen_at":"2026-11-20T03:00:47Z","employer_posted_date":null,"last_verified_at":"2026-11-20T03:00:47Z","board_verified":false,"closed_at":null,"days_open":0,"trust":{"level":"not_scored","repost_count":null,"flags":[],"days_open":0},"description":"1. This is a Noida based role: Do NOT apply if you are not from the NCR or are willing to relocate.\n\n2. This is a complete WFO opportunity: Do NOT apply if you are looking for a remote or hybrid work opportunity.\n\n3. The requirement is a minimum of 6 years of relevant experience: Do NOT apply if you have less than 6 years of hands-on, real-world relevant experience.\n\nPosition: Staff AI Engineer, Open-Source GenAI (R&D with Production Ownership)\n\nLocation: Noida\n\nDepartment: AI/ML\n\nEmployment Type: Full-time\n\nExperience: 6 - 8+ years\n\nAbout Gigaforce:\n\nGigaforce is a California-based InsurTech company delivering a next-generation, SaaS-based claims platform built for the Property and Casualty industry. We combine AI-powered models with deep domain expertise to streamline and accelerate subrogation and claims processing for insurers, recovery vendors, and other ecosystem participants, turning a traditionally fragmented claims lifecycle into an intelligent, end-to-end digital experience.\n\nOur AI products focus on subrogation, claims recovery, workflow automation, and decision support, including claim scoring, liability analysis, document preparation, demand generation, and agentic claims workflows. We build on open-source technology and run our own models, so our engineers work close to the model, not just behind an API.\n\nGigaforce was a finalist at InsurtechNY, was twice named a Top 50 company by the TiE Silicon Valley community, and was selected for the global accelerator program at Plug and Play Tech Center in Sunnyvale, California. Our platform is built with Java, Kotlin, Angular, Node.js, and a modern AI stack.\n\nWe are driven by curiosity, excellence, and inclusion. Our employees enjoy comprehensive medical benefits, equity participation, and generous paid time off. We are an equal opportunity employer and are proud to foster a diverse, equitable, and inclusive workplace.\n\nRole Overview:\n\nWe are looking for a Staff AI Engineer to lead the research and production hardening of open-source GenAI systems for insurance claims and subrogation. You will work directly with the AI Lead, who sets the overall architecture, and turn it into a concrete technical plan: which models to use, how to serve and evaluate them, and how to make multi-agent and document-intelligence workflows reliable enough for customers.\n\nThis is a hands-on role. You will run experiments on open models, fine-tune them where prompting is not enough, and own the path from prototype to production, while mentoring mid-level engineers who implement components alongside you. Much of our data is documents and text (claim files, adjuster notes, legal correspondence, scanned records), so retrieval, extraction, and document understanding sit at the center of the work.\n\nWe build on open-source models and self-host them for cost, control, and data privacy. Experience limited to hosted APIs (Gemini, OpenAI, Anthropic) will not be enough for this role.\n\nKey Responsibilities:\n\nOpen-Source Models and Fine-Tuning:\n\n- Evaluate and select open-source LLMs and embedding models (for example Llama, Qwen, Mistral, and Gemma families) against quality, latency, and GPU cost targets for specific claims tasks.\n\n- Fine-tune and adapt models using LoRA/QLoRA, instruction tuning, and distillation where prompting alone does not meet the quality bar.\n\n- Build and maintain training and evaluation datasets from claim documents, notes, and adjuster feedback.\n\nInference and Serving:\n\n- Deploy and operate self-hosted model inference using vLLM, SGLang, or TGI, including quantization, batching, and capacity planning.\n\n- Balance accuracy, throughput, latency, and cost across models and workloads, and define when to use smaller or larger models.\n\nDocument Intelligence and Retrieval:\n\n- Design and improve pipelines for document-heavy workflows: OCR, layout-aware parsing, classification, structured extraction, chunking for long claim files, embeddings, rerankers, and retrieval.\n\n- Build retrieval evaluation for noisy, scanned, and inconsistent insurance documents, and improve it through structured experimentation.\n\n- Extend retrieval and knowledge handling across lines of business (auto, property, and Workers' Comp) and across tenants in a multi-tenant SaaS setting.\n\nAgents, Evaluation, and Reliability:\n\n- Make multi-agent and tool-calling workflows work reliably on open models using structured output, constrained decoding, and guardrails.\n\n- Build evaluation harnesses that do not depend on a frontier API, using open judge models and human-labeled sets, and track quality across releases.\n\n- Design feedback loops that turn human-in-the-loop corrections into improved playbooks, prompts, and models over time.\n\n- Support human approval flows, auditability, and compliance requirements for agents that take action.\n\nProduction Engineering and Leadership:\n\n- Take AI use cases from prototype to production with the AI Lead's architecture, and define the production-readiness bar: testing, monitoring, logging, and rollout practices.\n\n- Expose model and agent functionality through APIs and MCP services in collaboration with platform, backend, and product teams.\n\n- Review code and designs, mentor mid-level engineers, and break larger initiatives into well-scoped work for the team.\n\nRequired Skills & Qualifications:\n\n- 6 - 8+ years of experience in AI, machine learning, or applied ML roles, with at least 1 - 2 years building LLM or RAG systems that reached production.\n\n- Hands-on experience running open-source LLMs, including self-hosted inference with vLLM, SGLang, TGI, or equivalent, and an understanding of quantization, batching, and GPU memory and cost trade-offs.\n\n- Practical experience fine-tuning open models with LoRA/QLoRA or similar methods using Hugging Face Transformers, PEFT, or TRL, including dataset preparation and evaluation.\n\n- Strong experience with document and text pipelines: OCR, layout-aware parsing, chunking, embeddings, rerankers, and structured information extraction on noisy, real-world documents.\n\n- Experience building and operating RAG systems with vector stores or retrieval systems such as FAISS, Qdrant, Weaviate, Milvus, pgvector, or Chroma, and measuring retrieval quality.\n\n- Strong Python and solid foundations in supervised ML, model evaluation, and experiment tracking, with PyTorch as the primary framework.\n\n- Experience designing evaluation methods for LLM applications, including building test sets, using open judge models, and incorporating human review.\n\n- Experience deploying AI services to production: APIs, containers, monitoring, logging, testing, and performance tuning.\n\n- Ability to turn an architecture set by a lead into a clear technical plan, push back with data when needed, and mentor other engineers.\n\n- A practical mindset toward shipping systems that deliver measurable business outcomes, not demos.\n\nPreferred Qualifications:\n\n- Experience in insurance, claims, legal, or other document-heavy workflow automation domains.\n\n- Experience with multi-agent or tool-calling systems built on open models, including constrained decoding or structured generation libraries.\n\n- Exposure to multimodal and vision-language models for document understanding.\n\n- Familiarity with tools such as LangChain, LlamaIndex, LangGraph, MLflow, Docker, or Kubernetes.\n\n- Experience with LLM guardrails, safety, and compliance for systems that act on behalf of users.\n\n- Experience with model distillation, synthetic data generation, or preference tuning.\n\n- Public GitHub work, open-source contributions, or technical writing.\n\nWhat Success Looks Like:\n\n- You take an AI use case from research to a production system running on open models, with clear quality, latency, and cost numbers.\n\n- You improve model and retrieval quality through structured experiments, and can show the gains on a stable evaluation set.\n\n- You raise the team's engineering bar: mid-level engineers ship reliable components faster because of your plans, reviews, and mentoring.\n\n- You push back on the architecture with evidence when needed, and then commit fully to the direction the AI Lead sets.\n\n- You are motivated by applying AI to high-value operational problems in insurance, not just building demos.\nSkills\nGenerative AI, LLM, Python, PyTorch, Agentic AI, RAG, NLP, VectorDB","description_format":"text","description_chars":8321,"description_truncated":false,"requirements":{"experience_years_min":6,"management_years_min":null,"team_size_min":null,"manages_managers":false,"education":null,"security_clearance":false,"languages":[]},"benefits":["Equity","Hybrid work"],"hiring_locations":[],"hiring_excludes":[],"relocation_offered":false,"industries":["Supply Chain & Logistics Software","InsurTech"],"lifecycle":[{"event":"open","at":"2026-11-20T04:45:15Z"}],"visa":[],"liveness":{"score":86,"band":"hot","label":"Hiring now","p_open":1,"p_active":0.86,"p_room":1,"age_days":0,"expected_fill_days":34,"reasons":["seen:0","win:early"],"computed_at":"2026-11-20T05:46:30Z"},"pay":{"stated_usd_annual":32157,"is_top_pay":false},"html_url":"https://alion.io/job/gigaforce-software-staff-ai-engineer-open-source-genai","json_url":"https://alion.io/job/gigaforce-software-staff-ai-engineer-open-source-genai.json","meta":{"generated_at":"2026-11-20T06:55:53Z","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":4972,"day_limit":null,"remaining_today":null,"minute_limit":300,"resets_at":"2026-11-21T00:00:00Z"}},"offers":[{"id":"company.slices","title":"One company in depth, by slice","status":"live","price":{"credits":0.1,"usd":0.01,"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":3800764},"rest":"https://alion.io/mcp/rest/get_company?id=3800764"},"human":"https://alion.io/catalog?offer=company.slices&for=job%2Fgigaforce-software-staff-ai-engineer-open-source-genai"},{"id":"market.stats","title":"A market slice: pay, demand and time to fill","status":"free","price":{"credits":0,"usd":0},"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%2Fgigaforce-software-staff-ai-engineer-open-source-genai"},{"id":"job.search","title":"Open jobs by role, technology, place, pay and visa","status":"live","price":{"credits":0.1,"usd":0.01},"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%2Fgigaforce-software-staff-ai-engineer-open-source-genai"},{"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/gigaforce-software-staff-ai-engineer-open-source-genai\", \"note\": \"what you need it for\"}"},"human":"https://alion.io/catalog?offer=company.verify&for=job%2Fgigaforce-software-staff-ai-engineer-open-source-genai"}],"agent":{"mcp":"https://alion.io/mcp/jobs","mcp_open":"https://alion.io/mcp/open","next":[{"tool":"get_job","arguments":{"id":4658812}},{"tool":"search_jobs","arguments":{"company":"gigaforce.io","detail":"compact"}},{"tool":"match_people","arguments":{"job_id":4658812},"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"}}