Anti-ATS Evaluatorv1.3

[ANALYSIS_REPORT] **HIGH** [3-Flash] The candidate's technical profile perfectly covers the LLM, RAG, and Python requirements, including libraries like LangChain. Although titled 'Mid-Level', the role covers the full pipeline from conception to deploy, which the candidate does autonomously. The 'Mechatronics' background satisfies the technical degree requirement easily. **Strengths:** LLM/RAG/LangChain, Python/SQL Mastery, Full-cycle project experience

[ANALYSIS_REPORT] **TOP** [3-Flash] This role's focus on 'industrial problems' and 'maintenance reports' using NLP is a direct hit for a Mechatronics Engineer who has managed industrial maintenance (Pronutri). The candidate's background in both hardware/control systems and modern NLP/LLM makes them a unique 'purple squirrel' for this specific vacancy. Mention the 4,000+ record cleanup in the maintenance context. **Strengths:** Mechatronics/Industrial background, NLP for maintenance data, End-to-end autonomy

[ANALYSIS_REPORT] **HIGH** [3-Flash] The candidate has the required LLM, NLP, and unstructured data experience at a senior level. The gaps are mainly platform-specific (Databricks, Spark) and MLOps tools. Highlighting the 2.9% classification variance achievement would demonstrate the 'validation and deployment' skills required for a senior role. **Strengths:** LLM/NLP extraction and classification, Python/SQL expertise, Senior autonomy **Critical Gaps:** Distributed processing (Spark/Databricks) **Missing Required:** Databricks, Spark

[ANALYSIS_REPORT] **LOW** [3-Flash] The job description lists 'Título de Doutor' (PhD) as a mandatory requirement. The candidate has a Bachelor's degree (Bacharelado). This is a hard filter in academic/research-heavy scholarship programs like Inova Talentos. **Strengths:** Python, Machine Learning, NLP **Critical Gaps:** PhD requirement not met **Missing Required:** Título de Doutor

[ANALYSIS_REPORT] **LOW** [Copilot: GPT5.1] This role expects a lead applied scientist with a strong record of IR/NLP research, published work, and leadership in applied research environments, which is not reflected in the candidate’s background. The candidate is more of a systems architect and AI/LLM practitioner than a traditional scientist with publications or formal research leadership. Bridging this gap would require a history of research outputs, IR/NLP system design at research depth, and demonstrated leadership of scientists. **Strengths:** Strong applied AI/LLM and systems engineering background, Ability to translate business problems into technical AI architectures, Experience collaborating with product and engineering stakeholders in delivery contexts **Missing Required:** Hands-on experience developing IR and NLP systems as a primary scientific authority, Proven experience scaling impact by leading others in an applied research environment, Master's degree plus equivalent research-heavy experience in a relevant discipline

[ANALYSIS_REPORT] **TOP** [3-Flash] The candidate matches the GenAI, NLP, and Node.js/JavaScript requirements, which is a rare combination for a Data Scientist. Their experience with scrapers and RAG fits the 'Conversational AI' and 'NLP' needs. Gaps include specific cloud tools (SageMaker, Watsonx) and some MLOps frameworks. **Strengths:** GenIA/NLP mastery, JavaScript/Node.js + Python, End-to-end solution delivery **Critical Gaps:** Specific enterprise Conversational AI platforms (Watson) **Missing Required:** Python/SQL/JS, LLM & RAG experience

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] Candidate has strong LLM pipeline experience (Claude/Gemini/OpenAI APIs, prompt engineering, LangChain listed), Python, SQL, and CI/CD — but lacks the core Data Science stack: no Pandas/NumPy/Scikit-learn demonstrated, no ML model training experience, no deep learning frameworks (PyTorch/TensorFlow), no AWS Bedrock experience, and no multi-agent systems built. The role demands a traditional ML/DS background with heavy statistical modeling that the candidate does not possess. To improve, candidate should highlight any statistical forecasting work from Supply Chain, add Pandas/NumPy to demonstrated skills, and build a project using AWS Bedrock with multi-agent architecture. **Strengths:** LLM API experience (Claude, Gemini, OpenAI), Prompt engineering with measurable results, Python and SQL proficiency **Critical Gaps:** No ML/DS model training experience, No deep learning framework experience **Missing Required:** 2+ years ML/DS experience, ML/DS libraries (Pandas, NumPy, Scikit-learn), AWS Bedrock, multi-agent systems

[ANALYSIS_REPORT] **HIGH** [Copilot: GPT5.1] This role is highly aligned with the candidate’s recent work running LLMs in production, designing RAG pipelines, and handling inference reliability on constrained infrastructure. The resume shows strong experience in LLM pipeline design, prompt engineering, and production deployment, though it lacks explicit mentions of specific ML frameworks and cloud stacks. To maximize ATS and human fit, the candidate should call out MLOps practices, any use of frameworks like PyTorch/TensorFlow, and cloud services used for hosting inference. **Strengths:** Hands-on experience supporting LLM inference and reliability in production, Strong understanding of NLP, prompt engineering, RAG, and evaluation for language models, Experience designing and operating full AI pipelines from data to deployment

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] Candidate has relevant LLM experience (RAG listed in skills, prompt engineering, Claude/Gemini APIs), Python, SQL, and the engineering degree qualifies under 'areas correlatas'. However, the role requires supervised/unsupervised ML model development, NLP beyond LLM APIs, deep learning, and data visualization tools — none of which the candidate demonstrates. The 'Assistente Virtual' and RAG optimization tasks partially align with candidate's RAG and prompt engineering experience. To improve, candidate should emphasize the statistical forecasting from supply chain as ML-adjacent experience and highlight the RAG/embedding/vector search skills more prominently. **Strengths:** LLM and GenAI experience with production pipelines, RAG architecture knowledge, Engineering degree (correlata) **Critical Gaps:** No ML model training/validation experience **Missing Required:** ML model development experience, NLP beyond LLM APIs

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] This role is heavily focused on LLM agents, RAG architecture, and prompt engineering — areas where the candidate has genuine production experience. Candidate has RAG (listed in skills with embeddings, vector search, chunking), LangChain, prompt engineering with statistical validation, Python, SQL, and REST API experience. The 5+ years DS/ML requirement is a major gap since candidate has 0 years with that title. Cloud experience is weak (only 'AWS fundamentals'). The 'ReAct / tool calling' agent pattern is missing. To improve, candidate should frame the market intelligence platform as an agent system and emphasize the RAG pipeline components explicitly. **Strengths:** RAG architecture (embeddings, vector search, chunking), Prompt engineering with measurable variance reduction, Production LLM pipeline in Go with Python **Missing Required:** 5+ years Data Science/ML experience

[ANALYSIS_REPORT] **MEDIUM** [Opencode: opus-4-6] This is the strongest match among the DS roles. Candidate has direct experience with: RAG (embeddings, vector search, chunking — listed in skills), LLM pipeline (Claude/Gemini/OpenAI APIs), prompt engineering with measurable results (26% to 2.9% variance), Python, SQL, PostgreSQL, Docker, Git, CI/CD, and LangChain. The engineering degree qualifies as 'areas relacionadas'. Key gaps are vector database specifics (FAISS/Pinecone/Weaviate/Chroma — candidate uses PostgreSQL for vector storage which is adjacent), ML/NLP beyond LLMs, and tool calling/agent orchestration patterns. To improve, candidate should specify which vector storage approach was used, mention tool calling if the orchestrator uses it, and add LlamaIndex to learning priorities. **Strengths:** Production RAG pipeline with embeddings and chunking, Prompt engineering with statistical validation, LangChain + multiple LLM API experience, Docker, Git, CI/CD, Cloud fundamentals **Missing Required:** Solid ML/NLP experience beyond LLM APIs, Vector database experience (specific tools)

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] This role requires proven Computer Vision experience — a completely different domain from the candidate's skillset. While the candidate has Python, SQL, and can build APIs, the core requirements (CV models, NLP model training, GCP, Flask/FastAPI, MLOps) represent fundamental gaps. Computer Vision is not inferable from any of the candidate's demonstrated work. The 'consultoria ou áreas de negócio' nice-to-have is a minor positive. To improve: this role is not a fit — the candidate would need months of CV project work to be competitive. **Strengths:** Python and SQL, Business domain experience (consultoria adjacent), API construction experience **Critical Gaps:** No Computer Vision experience whatsoever **Missing Required:** Computer Vision project experience, GCP familiarity, Flask/FastAPI, MLOps

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] This Data Engineering role requires Databricks on Azure, PySpark, Hadoop ecosystem, MongoDB, and heavy ETL pipeline experience — all of which the candidate lacks. While the candidate has Python, SQL, Git, API pipeline experience, and data quality work, the specific big data tooling (Databricks, PySpark, Hadoop) represents a critical gap in a different technology domain. The candidate's data pipeline work is at a much smaller scale than what this role demands. To improve: candidate would need to learn Databricks/PySpark and build ETL projects at scale. **Strengths:** Python and SQL, API-based pipeline construction, Data quality and validation experience **Critical Gaps:** No Big Data tooling experience (Databricks/PySpark/Hadoop) **Missing Required:** Databricks Azure, PySpark, MongoDB, Hadoop

[ANALYSIS_REPORT] **LOW** [Copilot: GPT5-mini] Strong systems and PostgreSQL familiarity plus Docker and RAG experience, but lacks the JD's core required FastAPI expertise, pgvector hands‑on experience, BigQuery and Kubernetes exposure. ATS will penalize missing explicit FastAPI/async/Pydantic and pgvector keywords despite relevant adjacent LLM/vector work. Improve by surfacing any FastAPI/async projects, explicit PostgreSQL performance work, BigQuery/K8s experience, and pgvector or similar vector-store usage on the resume/keywords. **Strengths:** PostgreSQL (listed in stack), Docker / Linux / container familiarity, LLM / RAG / embedding pipeline experience **Missing Required:** FastAPI (expert-level), pgvector, BigQuery, Kubernetes

[ANALYSIS_REPORT] **MEDIUM** [Opencode: opus-4-6] This is one of the better matches. The role is Backend + AI for PoCs/MVPs — very aligned with candidate's builder profile. Candidate has: Python, SQL, JavaScript (listed), PostgreSQL (relational DB optimization), Git/GitHub, VS Code, LLM API experience (advanced prompt development), and pipeline/orchestration experience. The candidate built production systems integrating LLMs with data pipelines. Gaps: no vector database (Pinecone specifically), no TensorFlow/PyTorch, and TypeScript is not explicitly listed (only JS). The PoC/MVP focus matches candidate's demonstrated rapid prototyping ability. To improve: add TypeScript to skills if familiar, mention vector search experience explicitly, and frame the orchestrator as demonstrating 'workflow orchestration and API integration'. **Strengths:** Production LLM pipeline with orchestration, Python + JavaScript + SQL proficiency, PoC/MVP builder mentality with rapid delivery, API integration and webhook experience **Missing Required:** TensorFlow/PyTorch (ML libraries), Vector database (Pinecone)

[ANALYSIS_REPORT] **LOW** [Copilot: GPT5-mini] Candidate has strong systems and Python experience but lacks numerous mandatory large-scale data platform technologies (ClickHouse, Kafka/Flink, Spark, Airflow, Terraform, ClickHouse/Elastic/Neo4j specific experience). Multiple required tool omissions trigger heavy deductions so the computed score collapses and effectively filters out in ATS. To improve, surface any experience with streaming, columnar stores, infra as code, and large-cluster ops, or target less specialized data-platform roles first. **Strengths:** Distributed systems architecture experience, Python and SQL programming, Experience building production-oriented orchestrators and monitoring dashboards **Missing Required:** ClickHouse, ElasticSearch, Spark/Kafka/Flink, Airflow, Terraform/Kubernetes

[ANALYSIS_REPORT] **LOW** [Copilot: GPT5-mini] Background in mechatronics gives domain transferability but the JD requires extensive, explicit OpenSCAD and parametric script experience which the resume does not show. ATS will see the mismatch in tool keywords and likely deprioritize this candidate. If interested, build and link OpenSCAD portfolio projects and state parametric/CAD experience prominently. **Strengths:** Mechatronics engineering background, Control-systems and practical prototyping experience, Strong written technical clarity **Missing Required:** Extensive OpenSCAD project experience, Parametric module development

[ANALYSIS_REPORT] **LOW** [Copilot: GPT5-mini] Mechatronics background and systems modeling provide some transferable capabilities, but the role demands energy engineering, PV/storage, and domain-specific modeling experience that aren’t present on the resume. ATS will flag the absence of the explicit energy/system-design keywords and domain tools. To improve, add any energy/renewables modeling coursework, validation/rubric evaluation examples, and specific experience with PV/storage/heat-pump modeling or energy-system tools. **Strengths:** Control-systems/mechatronics engineering foundation, Experience with simulation and rigorous technical documentation, Automation and evaluation experience (AI model rubric familiarity) **Missing Required:** Energy engineering / renewable energy modeling experience, Experience with energy-system performance modeling

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] This role requires an active CREA registration, a Master's degree in specialized mechanical engineering topics (rotor dynamics, finite elements, CFD), and 100% presencial in Campinas/SP. The candidate has a Mechatronics Engineering degree but no CREA ativo mentioned, no Master's degree, and the domain (asset management for rotating machinery) is completely outside the candidate's trajectory. The IoT/sensor and predictive modeling aspects are tangentially related but the core requirements are hard engineering certifications. This is a hard filter failure. **Strengths:** Mechatronics Engineering degree, Python knowledge, English advanced **Critical Gaps:** No CREA ativo (legally required), No Master's degree (required), No mechanical asset management experience **Missing Required:** CREA ativo, Master's degree in specified engineering topics, 100% presencial Campinas

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] This role requires a Master's in Engineering (Mechanical, Electrical, Aerospace, Industrial, or Automotive), prior automotive sector experience, deep knowledge of vehicle propulsion systems (ICE, transmissions, batteries, electric motors), hybrid control strategies, ECMS, MATLAB/Simulink, Stateflow, and GT-Suite. The candidate has none of these domain-specific requirements. While the candidate has an engineering degree and control systems background from Mechatronics, the specific automotive powertrain domain expertise is completely absent. This is a fundamentally different engineering discipline. **Strengths:** Engineering degree with control systems background, English proficiency, Git experience **Critical Gaps:** No automotive powertrain experience, No Master's degree, No MATLAB/Simulink **Missing Required:** Master's in Engineering, Automotive sector experience, MATLAB/Simulink, Vehicle propulsion knowledge

[ANALYSIS_REPORT] **MEDIUM** [Copilot: GPT5-mini] Solid quantitative and modeling background plus management experience align with the modelling specialist responsibilities, but the JD seeks domain-specific power‑market modelling and (likely) a PhD – both of which are missing. Resume strengths in modeling, Python, and managing junior colleagues help, but the absence of power-market experience and higher academic credential reduce ATS ranking. Improve by highlighting any market/power-related analysis, optimization/MIP work, and leadership of modelling projects; call out Python modelling libraries and relevant commercial model outputs. **Strengths:** Quantitative modeling experience (trading, forecasting), Python programming skills, Management/mentoring experience and commercial modelling exposure **Missing Required:** Power market modelling experience, Relevant PhD (if mandatory)

[ANALYSIS_REPORT] **LOW** [Opencode: opus-4-6] This role requires a degree in Hydraulic, Environmental, or Civil Engineering, plus deep domain knowledge in hydrology, hydraulics, water quality, hydrological modeling tools (HEC-HMS, HEC-RAS), and GIS (QGIS). While the candidate has strong Python, automation, and analytical skills that partially match the computational aspects, the core domain (hydrology, water systems, environmental engineering) is entirely outside the candidate's background. The engineering degree in Mechatronics does not qualify as 'áreas correlatas' for hydrology. This is a domain mismatch. **Strengths:** Python proficiency, Automation and scripting ability, Analytical reasoning **Critical Gaps:** No hydrology/environmental engineering background, Wrong engineering discipline **Missing Required:** Engineering degree in Hydraulic/Environmental/Civil, Hydrology knowledge, HEC-HMS/HEC-RAS, GIS tools

[ANALYSIS_REPORT] **MEDIUM** [Copilot: GPT5-mini] Relevant AI/LLM pipeline design, evaluation mindset and systems thinking are transferable to RL environment design, but the JD expects explicit RL environment/reward engineering experience. ATS will penalize missing direct RL-environment keywords (reward functions, verifiers, synthetic task generation) despite strong adjacent signals. To improve, surface any RL experiments, reward-function design, environment interfaces, or synthetic-data pipelines and include quantitative evaluation outcomes. **Strengths:** LLM/RAG and pipeline design experience, Strong systems engineering and validation mindset, Experience building quantitative evaluation and monitoring systems **Missing Required:** 4–5 years building deep learning or RL systems (direct RL env experience)

[ANALYSIS_REPORT] **LOW** [Copilot: GPT5-mini] Mechatronics education is relevant but the JD requires deep experience with NX, CAE, and specific vehicle/assembly modeling skills not shown on the resume. ATS will deprioritize due to absent NX/CAE keyword matches and part-level NX Analysis experience. If interested, add explicit NX, CAE, and vehicle-systems simulation projects and certified tool experience. **Strengths:** Mechatronics foundation and systems-level thinking, Experience documenting technical specifications, Hands-on engineering and prototyping mindset **Missing Required:** NX modeling and CAE experience specific to vehicle subsystem design

[ANALYSIS_REPORT] **LOW** [Copilot: GPT5-mini] The role needs specialist experience in Gaussian splatting, neural reconstruction, and real-time engine integration (Unity/Unreal) — none of which appear on the resume. ATS will treat this as a categorical mismatch to 3D reconstruction/game-engine specialties. Build demonstrable projects (NeRF/Gaussian splatting pipelines, Unity/Unreal integrations) before applying. **Strengths:** Strong systems engineering and prototype delivery, Experience with simulation and robotics foundations (mechatronics) **Missing Required:** Hands-on neural reconstruction and game-engine integration experience