Dodge Construction Network (Dodge) is looking for a Senior ML & Automation Engineer to design and operate AI-powered pipelines that transform how we acquire, enrich, and validate construction project intelligence. This role blends machine learning, document automation, and conversational AI — spanning internal operational tooling today and expanding into automated outbound AI communications to the field. You will partner closely with Data Engineering and Operations to move pipelines from prototype to production and maintain them rigorously at scale.This is a full-time position and reports directly to the Director, Data Operations.Preferred LocationThis is a remote, home-office based role and candidates located in India will be considered. Travel RequirementsOccasional travel to the Kochi office is required.Essential Functions Document Intelligence & ExtractionDesign, develop, and evaluate machine learning models to automate data enrichment, classification, and validation across structured and unstructured project documentsImplement OCR, NLP, and layout recognition pipelines to extract metadata, contacts, deadlines, and technical requirements from plan sets, specifications, and bid documentsBuild Python-based classification microservices to categorize documents by type and extract structured fields (e.g., bid dates, scope of work, discipline sheets, spec sections)Integrate LLM APIs (AWS Bedrock, Anthropic Claude, or equivalent) for intelligent extraction and classification tasks; optimize prompts and model calls for accuracy and cost efficiencyOwn model performance monitoring for one or more document domains — tracking accuracy drift, false positives, and cost per-document over timeConversational AI & AWS ConnectDesign and implement conversational AI solutions using Amazon Connect and Amazon Lex, including contact flows, IVR design, and agent assist integrations for internal operational toolingBuild and iterate on automated outbound AI calling workflows to collect project updates from contractors, subcontractors, and field contacts — capturing structured responses and routing them into data pipelinesEnsure all outbound communication automation is implemented in compliance with applicable regulations (TCPA, Do Not Call rules, B2B communication standards); partner with Legal and Compliance ahead of any production deploymentDesign conversation scripts with dynamic branching logic and fallback handling; continuously improve containment rates and data capture qualityMonitor call performance, intent recognition accuracy, and fallback rates; iterate on Lex models and contact flows based on outcomesData Pipelines & Entity ResolutionBuild pipelines that integrate scraped and API-sourced project data with external enrichment sources (ZoomInfo, LinkedIn, government open data APIs) to enrich company, contact, and project recordsImplement entity resolution and record deduplication logic — including fuzzy name matching, license number anchoring, and cross-source reconciliation — to maintain a clean entity masterDevelop automation scripts and microservices to reduce manual effort in project matching, contact discovery, and quality checksCollaborate with Data Engineers to ensure ML pipelines integrate seamlessly with existing data warehouses (Redshift) and meet latency and cost targetsPartner with data specialists to design feedback loops that validate and continuously improve model outputsEducation RequirementBachelor’s degree in a related field or equivalent education and work experience.Required Experience, Knowledge and Skills5+ years of experience in machine learning, automation engineering, or a closely related disciplineProficiency in Python with hands-on experience using ML libraries (scikit-learn, spaCy, TensorFlow, or PyTorch) and production API integrationHands-on experience with OCR frameworks — Tesseract, PaddleOCR, AWS Textract, or Google Document AIDemonstrated experience implementing AWS Connect solutions — including contact flow design, Amazon Lex bot development, and IVR configurationPractical knowledge of LLM APIs (AWS Bedrock, OpenAI, Anthropic, or equivalent) for production extraction or classification workloadsFamiliarity with document layout analysis tools (LayoutLM, Donut, DocTR, or similar)Strong knowledge of entity extraction, NER, regex-based parsing, and rules-based approachesExperience with entity resolution, deduplication, or fuzzy record matching at scaleStrong knowledge of data pipelines and ETL frameworks; experience deploying and monitoring ML models in productionSolid understanding of relational databases and SQL; experience with large-scale warehouses (Redshift, Snowflake, or similar)Awareness of outbound communication compliance (TCPA, Do Not Call regulations) in automated or AI-driven calling contextsStrong problem-solving skills with the ability to translate operational business needs into ML and automation solutionsPreferred Experience, Knowledge and SkillsExperience with outbound AI calling automation — proactive conversational agents, dynamic call scripting, and structured data capture from voice interactionsFamiliarity with government and open data APIs (Socrata, ArcGIS, Legistar, Granicus, or similar) for large-scale public data ingestionExperience with AWS analytics and ML services — SageMaker, Comprehend, Lambda, Step FunctionsAwareness of inference cost optimization and batch processing strategies for LLM-powered production pipelinesExposure to CI/CD and MLOps tooling — MLflow, Git, Docker, KubernetesPrior experience with sales intelligence or contact enrichment data (ZoomInfo, LinkedIn, or comparable sources)Experience in Agile delivery environments using Jira or ConfluenceAbout Dodge Construction Network Dodge Construction Network exists to deliver the comprehensive data and connections the construction industry needs to build thriving communities. Our legacy is deeply rooted in empowering our customers with transformative insights, igniting their journey towards unparalleled business expansion and success. We serve decision-makers who seek reliable growth and who value relationships built on trust and quality. By combining our proprietary data with cutting-edge software, we deliver to our customers the essential intelligence needed to excel within their respective landscapes. We propel the construction industry forward by transforming data into tangible guidance, driving unparalleled advancement. Dodge is the catalyst for modern construction.
