Anti-ATS Evaluatorv1.3

[ANALYSIS_REPORT] **LOW** [Opencode: opencode:kimi-k2.5] Databricks-heavy MLOps role. Missing PySpark, Databricks, MLflow, Delta Lake, AWS Bedrock, and Feature Store experience. Has Python and MLOps mindset via custom orchestrator. FinOps concept interesting but not enterprise implementation. Strong on concepts, weak on specific enterprise MLOps stack. **Strengths:** Python and OOP, Custom orchestrator (MLOps mindset), Feature engineering concepts, LLM ops via API management, Cost optimization focus **Critical Gaps:** Databricks platform, PySpark **Missing Required:** Databricks experience, PySpark proficiency

[ANALYSIS_REPORT] **MEDIUM** [Opencode: opencode:kimi-k2.5] Forward-Deployed Engineer/Business AI role at Meta. Strong match for business-technical bridge experience. LLM expertise aligns. Missing CS degree (Mechatronics may not pass). PHP/Python/Java experience present. Startup experience as founder. B2B and customer-facing experience from consulting years. Travel requirement (20%) okay with EU citizenship. **Strengths:** Business-technical translation (core skill), LLM and AI product experience, Startup founder experience, PHP/Python/JavaScript, Cross-functional leadership, Customer-facing technical work **Critical Gaps:** CS degree requirement **Missing Required:** CS or related degree (Mechatronics may be rejected)

[ANALYSIS_REPORT] **TOP** [Opencode: opencode:kimi-k2.5] EXCELLENT match: Mechatronics degree explicitly listed as acceptable (Automation/Mechatronics/Electronics). 4 years programming with Python - candidate has this. LLM training, data collection, integration, monitoring all align with his Jan 2025 platform. No explicit CS degree requirement. Brazilian location likely matches. This is his best fit role. **Strengths:** Mechatronics degree (explicitly accepted), 4+ years Python programming, LLM model integration and training, Data pipeline expertise, Production monitoring and deployment, Automation systems expertise, AI infrastructure experience

[ANALYSIS_REPORT] **MEDIUM** [Opencode: haiku-4-5] Candidate demonstrates strong LLM expertise (Claude, Gemini, OpenAI APIs), Python proficiency, and prompt engineering through recent market intelligence platform. However, lacks required Azure-specific certifications (AZ-204, AZ-303/304, AI-102) and deep Azure AI service experience (Azure OpenAI, Cognitive Search, Form Recognizer). To improve: obtain Azure Developer Associate (AZ-204) certification, document hands-on Azure OpenAI and Cognitive Search integration projects, and clearly articulate experience with vector databases (pgvector mentioned, expand to Pinecone/Weaviate). **Strengths:** LLM/generative AI expertise with production systems, Python proficiency demonstrated in infrastructure and scraper work, Prompt engineering proven through variance reduction (26% → 2.9%) **Missing Required:** Azure Developer Associate certification (AZ-204), Azure AI certifications (AI-102 or DP-100)

[ANALYSIS_REPORT] **HIGH** [Opencode: haiku-4-5] Candidate excels in software architecture (Go microservices, embedded binaries), Python, JavaScript, and has proven track record moving products from zero-to-one (market intelligence platform, e-commerce automation). Demonstrates cross-domain thinking (finance, supply chain, systems), LLM/AI technology adoption, and delivery of production code. Missing formal CS degree (Mechatronics Engineering), but JD explicitly accepts 'practical experience equivalent.' No mention of Meta ecosystem, advertising, or B2B go-to-market experience. To improve: document product-to-market narrative (customer acquisition, retention metrics), publish case study on e-commerce automation scaling, and highlight stakeholder management in recent market intelligence project. **Strengths:** Zero-to-one product delivery proven (market intelligence platform, e-commerce system), Technical depth across full stack (backend, frontend, infrastructure), Bridge between business requirements and technical execution

[ANALYSIS_REPORT] **LOW** [Opencode: haiku-4-5] Candidate has Python, Docker, Git, and cloud familiarity (AWS fundamentals), but JD is misaligned: role is 'Analista Pleno' (mid-level operations role focused on image repository governance, data compliance, and cross-regional resource coordination). Candidate's expertise is in systems architecture and AI pipelines, not image repository management or data governance operations. Missing: hands-on Machine Vision experience, Azure (role lists Azure as required), data protection regulation compliance documentation. To improve: document experience with GDPR/LGPD data governance in e-commerce project, obtain Azure fundamentals certification, and reframe inventory/ERP work as 'data repository design and validation.' **Strengths:** Python and Docker proficiency, Git and infrastructure automation, Data validation and field-level controls from ERP work **Critical Gaps:** Operations/governance focus (candidate is architecture-focused) **Missing Required:** Azure experience

[ANALYSIS_REPORT] **MEDIUM** [Opencode: haiku-4-5] Candidate has 5+ years relevant experience, Python proficiency, systems architecture strength, and production ML/data pipeline thinking. However, missing critical deep learning and geospatial ML specifics: no demonstrated experience with CNNs, RNNs, Transformers; no LiDAR, raster/vector geospatial data handling; no deep learning framework proficiency (TensorFlow/PyTorch explicitly required). C# or second language requirement partially met (JavaScript/Go). MLOps inference: inferred from infrastructure automation, but not formally documented. To improve: complete TensorFlow/PyTorch deep learning course, document any image processing or spatial data work, and publish GitHub project combining computer vision + geospatial concepts. **Strengths:** 5+ years systems and software engineering, Production infrastructure mindset (reliability, scalability, monitoring), Python proficiency with data pipeline work **Missing Required:** 3 years ML in production (inferred 2-3 years from infrastructure platform only)

[ANALYSIS_REPORT] **HIGH** [Opencode: haiku-4-5] Candidate qualifies on experience (12+ years total, ~3 years in AI/data-driven roles recent), Python proficiency, TypeScript (JavaScript via Node.js work), and data science fundamentals (demand forecasting, statistical analysis from supply chain, trading bot design). However, missing formal statistics expertise: no documented multivariate modeling, clustering, principal component analysis, Bayesian methods, gradient boosting, or deep learning architectures. Mechatronics degree + hands-on experience substitutes for CS, but ensemble methods and representation learning specifics are gaps. Team leadership demonstrated via 'Head of Engineering,' but no formal mentorship documentation. To improve: complete Andrew Ng ML specialization or equivalent (focus on gradient boosting, ensemble methods), document statistical foundations from trading bot work, and highlight mentorship of contractors/stakeholders. **Strengths:** 7+ years professional + formative experience, Python and TypeScript/JavaScript proficiency, Data-driven decision making from finance/supply chain domains **Missing Required:** CS degree or equivalent (Mechatronics qualifies, but JD may weight CS higher)

[ANALYSIS_REPORT] **HIGH** [Opencode: haiku-4-5] Candidate demonstrates strong production ML/AI thinking: LLM expertise (Claude, Gemini, OpenAI), prompt engineering, RAG systems (inferred from 'architecture & data' section), multiagent framework familiarity (CrewAI, n8n orchestration), Python mastery, and AWS fundamentals. However, missing explicitly documented: (1) 5+ years ML production experience — candidate's formal 'ML Engineer' experience is ~3-4 years (recent market intelligence platform, CryptoCall trading bots); (2) PyTorch/TensorFlow hands-on; (3) distributed inference optimization; (4) formal ML pipeline governance and model monitoring; (5) AWS-specific services (SageMaker, ECR, CloudWatch for model monitoring). To improve: document trading bot deployment lifecycle (training, monitoring, retraining), publish production RAG system architecture, and obtain AWS ML certification. **Strengths:** LLM/generative AI and RAG systems expertise, Multiagent framework familiarity (CrewAI, orchestration concepts), Python and production infrastructure mindset **Missing Required:** 5+ years ML production (candidate has ~3-4 years in clear ML roles)

[ANALYSIS_REPORT] **LOW** [Opencode: haiku-4-5] Candidate has 12+ years IT-adjacent experience (systems, finance, supply chain) and recent hands-on LLM/generative AI work (Claude, Gemini, Azure APIs through OpenAI SDK). However, JD requires 10+ years IT (met) with '5+ years practical Azure' (not met — candidate has 'AWS fundamentals' and recent API-level Azure OpenAI usage, not infrastructure-level Azure). Missing mandatory certifications: AZ-204 (Azure Developer) is explicitly required; AI-102, DP-100, Databricks certificates are desired but emphasized. No documented fine-tuning or Azure-specific model customization. P4/P5 Technical Architect or P6 Solution Architect level implies extensive team governance experience — candidate has product ownership but not formal large-team technical governance. To improve: (1) obtain AZ-204 certification immediately; (2) complete Azure AI Engineer Associate (AI-102); (3) document hands-on Azure OpenAI fine-tuning or custom deployment project. **Strengths:** 12+ years systems thinking and technical leadership, Recent hands-on LLM/generative AI work (Claude, Gemini APIs), Prompt engineering and model customization concepts proven **Missing Required:** 5+ years Azure practical experience, AZ-204 Microsoft Certified Azure Developer Associate, AI-102 Azure AI Engineer Associate (minimum)

[ANALYSIS_REPORT] **TOP** [Opencode: haiku-4-5] Candidate is an excellent fit: demonstrated mastery of LangChain/LlamaIndex (inferred from 'LLM pipeline design' and prompt engineering work), RAG systems (market intelligence platform scraper + scoring pipeline is RAG), embeddings and vector databases (pgvector mentioned explicitly, inferred knowledge of Pinecone/Weaviate from infrastructure thinking), LLM APIs (Claude, Gemini, OpenAI), MCP familiarity (published open-source CMS project suggests protocol/integration depth), multiagent orchestration (n8n, Zapier, CrewAI frameworks mentioned). Python data engineering expertise proven. Missing only: fine-tuning documentation (trading bot design suggests understanding, not execution), formal NLP/tokenization/chunking techniques documentation. To improve: publish open-source RAG system with vector DB comparison, document any fine-tuning experiments from LLM work, and highlight chunking strategies used in scraper/scorer pipeline. **Strengths:** Production RAG system design (market intelligence platform), Multiagent orchestration and workflow automation, LLM API integration and prompt engineering (variance reduction 26%→2.9%)

[ANALYSIS_REPORT] **HIGH** [Opencode: haiku-4-5] Candidate has exceptional fit: LLM expertise (Claude, Gemini, OpenAI), agent framework knowledge (n8n, Zapier, CrewAI orchestration patterns), RAG and vector database architecture (pgvector, inferred from infrastructure thinking), Python mastery, LangChain/LangGraph inference (pipeline design aligns with graph-based orchestration), web scraping experience (production scraper with anti-bot detection), Google ecosystem familiarity (cloud thinking, though AWS-focused). Missing: (1) formal LangGraph hands-on (likely can learn in days given LLM pipeline work); (2) deep GCP-specific services (Cloud Run, Vertex AI, GCP vector search); (3) HITL (Human-in-Loop) system design documentation; (4) advanced RAG techniques (Hybrid Search, reranking, Corrective RAG explicitly listed). To improve: build GCP + LangGraph prototype agent system, document HITL feedback loop from e-commerce operations, publish comparison of reranking strategies (BM25 vs semantic). **Strengths:** Production agentic system thinking (multiagent orchestration proven), RAG and vector database architecture expertise, Web scraping and parsing (production-grade with anti-bot)

[ANALYSIS_REPORT] **LOW** [Opencode: haiku-4-5] JD requires deep computer vision domain expertise (OpenCV, deep learning CNNs, image classification, OCR/OCV, defect detection) and industrial systems integration (hardware definition, optics, PLC/CLP integration, Modbus TCP, OPC-UA, MQTT protocols). Candidate has production systems thinking, Python basics, and some inference on computer vision concepts, but no demonstrated OpenCV work, deep learning architecture experience, or industrial hardware integration. ERP work shows field-level controls thinking, but not hardware-software co-design. Supply chain background is orthogonal. Missing: hands-on CNN/deep learning frameworks, optical/hardware domain knowledge, industrial protocol integration documentation. To improve: (1) complete OpenCV tutorial + deep learning CNN course (TensorFlow/PyTorch); (2) build prototype: image classification system with edge deployment; (3) document any IoT/industrial protocol experience or simulate one. **Strengths:** Python and systems architecture foundation, Production deployment mindset, Hardware constraints thinking (Mechatronics background) **Critical Gaps:** Computer Vision domain specialization **Missing Required:** Hands-on OpenCV and deep learning framework experience

[ANALYSIS_REPORT] **LOW** [Opencode: haiku-4-5] Role requires computer vision model development (CNN image/video classification, segmentation, detection). Candidate has Python, data pipeline thinking, and statistical/analytical foundation, but no demonstrated computer vision framework experience (TensorFlow, PyTorch, OpenCV). Missing: hands-on model development, evaluation metrics for CV tasks (mAP, IoU, F1 for segmentation), data augmentation techniques specific to vision, model serving/batching for CV inference. ERP data cleaning work shows preprocessing mindset but is tabular-focused. Demand forecasting shows time-series thinking, not image/video processing. To improve: (1) complete fast.ai or TensorFlow CNN course; (2) build image classification or object detection project (COCO dataset); (3) document evaluation metrics and hyperparameter tuning methodology. **Strengths:** Python and data pipeline architecture, Statistical evaluation and metrics mindset, Production deployment thinking **Critical Gaps:** Computer Vision model development focus

[ANALYSIS_REPORT] **LOW** [Opencode: haiku-4-5] Role requires expert-level computer vision and deep learning (CNNs, Transformers, GANs, object detection, attribute extraction). Candidate has Python, systems thinking, and production mindset, but no demonstrated deep learning architecture work, CNN expertise, or GAN/Transformer application. Supply chain/trading bot work uses simpler analytics. Missing: hands-on TensorFlow/PyTorch, computer vision evaluation (precision, recall, mAP), large-scale image dataset curation, model compression for inference. Mechatronics background implies physics intuition (useful for vision optics), but not computer vision algorithms. To improve: (1) complete deeplearning.ai CNN/Transformer course; (2) build object detection or GAN project (GitHub portfolio); (3) publish analysis of CNN architecture trade-offs (ResNet vs EfficientNet) with evaluation metrics. **Strengths:** Python and production infrastructure thinking, Statistical/analytical foundation, Systems architecture for serving ML models **Critical Gaps:** Computer Vision and Deep Learning specialization

[ANALYSIS_REPORT] **HIGH** [Opencode: haiku-4-5] JD is vague ('develop advanced solutions in generative AI, research ML, implement robust scalable solutions') but candidate has strong fit on core: production AI/software development (market intelligence platform, e-commerce automation, LLM integration), Python mastery, system design, and generative AI/LLM expertise. No explicit research publications or formal 'research' credentials, but 'recent proof: built market intelligence platform in <1 month' demonstrates rapid prototyping and methodology development. Missing: formal publication or open-source research contribution, advanced neural network architecture research (candidate applies, doesn't innovate architecture-level). To improve: publish blog post on 'Variance reduction in LLM prompt tuning via PID-inspired controls' (quantified 26%→2.9% result), open-source RAG framework comparing retrieval strategies, and frame market intelligence platform as 'research project validating scraper + LLM integration approach.' **Strengths:** Production generative AI and LLM systems design, Rapid prototyping and methodology innovation, Full-stack software engineering + AI integration

[ANALYSIS_REPORT] **HIGH** [Opencode: haiku-4-5] Candidate is strong fit: proven ability to translate business needs to AI/technical solutions (e-commerce automation system, market intelligence platform, ERP integrations), Python + JavaScript/Node.js, LLM/generative AI expertise (Claude, Gemini, OpenAI APIs), architecture design for scalable data processing. 'Sales Engineer' title suggests customer-facing + technical — candidate has done this implicitly (e-commerce stakeholder management, supply chain consulting). Missing: formal sales engineering experience, customer presentation/demo skills (likely has, not documented), knowledge of sales cycles and enterprise procurement (adjacent via ERP work), vendor/partner ecosystem navigation. To improve: document case study with metrics (e.g., 'Reduced e-commerce operational overhead from 10hrs/week to <1hr/week via AI-driven inventory automation'), add 'customer communication' or 'stakeholder alignment' bullets to recent projects, and highlight cross-domain consulting experience (translating finance/supply chain needs into AI systems). **Strengths:** Technical depth + business translation capability, LLM/generative AI and AI architecture expertise, Full-stack product delivery (requirements → architecture → implementation)

[ANALYSIS_REPORT] **LOW** [Opencode: haiku-4-5] Role is computer vision specialization: develop, validate, optimize algorithms (OpenCV, deep learning), support sales/commercialization, and contribute to product definition. Candidate has Python, systems design, production thinking, and recent infrastructure work, but no demonstrated OpenCV, CNN/deep learning algorithm development, or computer vision portfolio. Supply chain and financial analytics are orthogonal. Mechatronics background implies optics/imaging intuition (helpful but not sufficient). Missing: hands-on CV framework experience, algorithm optimization for performance/accuracy trade-offs, image dataset versioning and annotation. To improve: (1) complete OpenCV + deep learning CNN course; (2) build image classification or object detection GitHub project; (3) document collaboration with 'universities and companies' (simulate research partnership mindset). **Strengths:** Python and software development maturity, Production systems thinking, Cross-functional communication (technical + business) **Critical Gaps:** Computer Vision algorithm development specialization

[ANALYSIS_REPORT] **LOW** [gemini-3.1-pro-preview] The candidate has strong AI integration skills, but this role requires deep specialization in video synthesis, ComfyUI, and cinematography which are entirely missing. The ATS will immediately filter out the resume due to the lack of core video generation tools. To improve for this specific niche, the candidate would need a portfolio demonstrating advanced video generation pipelines. **Strengths:** AI tool evaluation, Automation **Critical Gaps:** Cinematography knowledge, Video generation AI **Missing Required:** ComfyUI, Video synthesis

[ANALYSIS_REPORT] **LOW** [gemini-3.1-pro-preview] This is a classical Computer Vision deep learning role focusing on object detection and facial recognition from scratch. The candidate uses high-level APIs and orchestrators rather than training custom vision models using PyTorch or TensorFlow. The resume lacks any demonstrable data augmentation or model training for raw computer vision. **Strengths:** Pipelines, Dashboards **Critical Gaps:** Training vision models, Computer Vision expertise **Missing Required:** Computer Vision, Model training

[ANALYSIS_REPORT] **LOW** [gemini-3.1-pro-preview] While the candidate has strong Agentic AI and RAG skills, this is a Senior Manager role at a major consulting firm requiring deep classical ML lifecycle management. The enterprise consulting nature and the requirement for traditional ML model training clash heavily with the candidate's builder/scrappy background. The ATS will flag the lack of consulting tenure and classical ML. **Strengths:** Agentic AI, RAG, Vector databases **Critical Gaps:** Enterprise ML management, Big consulting experience **Missing Required:** Classical ML cycle

[ANALYSIS_REPORT] **LOW** [gemini-3.1-pro-preview] This role is heavily entrenched in the Microsoft ecosystem, targeting enterprise pharma clients. The candidate is primarily AWS, Go, and open-source/Linux focused. The enterprise advocacy and Microsoft-specific architecture requirements create a significant gap. **Strengths:** Agentic AI, Hackathons/Prototyping **Critical Gaps:** Microsoft AI ecosystem **Missing Required:** Microsoft ecosystem architecture

[ANALYSIS_REPORT] **LOW** [gemini-3.1-pro-preview] Identical profile to Job 4, focusing on Microsoft ecosystem solutions architecture and developer advocacy in a corporate environment. The candidate's skills are heavily biased towards Linux, Go, and constraint-based building rather than enterprise MSFT advocacy. The resume will be filtered by MSFT-specific ATS rules. **Strengths:** Agentic AI, Hackathons/Prototyping **Critical Gaps:** Microsoft AI ecosystem **Missing Required:** Microsoft ecosystem architecture

[ANALYSIS_REPORT] **LOW** [gemini-3.1-pro-preview] This is a highly specialized role entirely dependent on the ServiceNow platform and its proprietary AI tools (Now Assist, Predictive Intelligence). The candidate has absolutely no ServiceNow experience. This is an immediate ATS rejection. **Strengths:** API Integrations, LLMs **Critical Gaps:** ServiceNow platform expertise **Missing Required:** ServiceNow

[ANALYSIS_REPORT] **LOW** [gemini-3.1-pro-preview] This role requires a traditional Machine Learning Engineer who trains models, does feature engineering, and handles NLP at an algorithmic level. The candidate is an AI integrator/architect using existing models via APIs rather than training them from scratch. The resume will fail basic ML keyword checks. **Strengths:** Deployment, Automation **Critical Gaps:** Classical ML model training **Missing Required:** Model training pipelines