Sr. Machine Learning Engineer Resume

Starting...

Sr. Machine Learning Engineer Resume Leo Pitts Sr. Machine Learning Engineer (512) 555-0184 leo.pitts.ml@example.com linkedin.com/example/leopitts Austin, TX 78701 STRENGTHS Research-to-production judgment Turned uncertain model ideas into tested PyTorch systems; teams gained confidence from clear experiments and dependable releases. Debug-first engineering Traced stubborn multi-GPU failures through gradients and communication; peers sought help on issues others could not isolate. Evaluation discipline Built benchmark suites separating capability gains from prompt effects; research partners gained clearer evidence for decisions. Collaborative technical leadership Coordinated engineers and stakeholders across research, hardware, and product; direct explanations earned executive trust. Practical model optimization Balanced quantization, latency, context, and capability for edge agents; runtime partners valued grounded engineering trade-offs. SKILLS Python LLM architectures Model training Evaluation frameworks Benchmark design Agent harnesses Context engineering Agent memory Tool use Post-training pipelines Supervised fine tuning Reinforcement learning Reward modeling GPU optimization Technical leadership LANGUAGES English Native 40 Spanish Intermediate 20 MY CAREER 15.9 Years Staff Machine Learning Engineer at Northstar Computing Group (3.2 Years) Principal Machine Learning Engineer at Redwood AI Systems (3.4 Years) Senior Machine Learning Engineer at Blue Mesa Research (2.9 Years) Machine Learning Engineer at VectorForge Analytics (3.4 Years) Software Engineer at HarborPoint Software (3 Years) SUMMARY Senior machine learning engineer and research-focused data scientist with 16+ years of software development experience and 13+ years delivering language-model research, data science, and production ML infrastructure. Builds evaluation frameworks , agent harnesses, post-training pipelines, reinforcement-learning environments, reward models, and resource-aware deployments. Brings hands-on command of Python, LLM architectures, supervised fine tuning, reinforcement learning, distributed training, PyTorch, CUDA, and model optimization across cloud and edge environments. Leads ambiguous research programs from hypothesis through reproducible experiments, checkpoint validation, deployment, and production support. Known for isolating numerical and systems issues, improving GPU utilization, mentoring engineers, and communicating research findings, architectural decisions, and trade-offs clearly across technical and non-technical teams. Eager to contribute this blend of rigorous research and practical engineering toward reliable agentic applications. EXPERIENCE Staff Machine Learning Engineer Northstar Computing Group July 2023 - Present Austin, TX Own research engineering for language-model post-training and agentic AI programs, coordinating six engineers across evaluation, data, training, runtime, and edge deployment workstreams. Translate research goals into reliable systems that improve model capability, alignment quality, and practical agent performance. Led benchmark development for tool use and reasoning, giving teams consistent checkpoint comparisons. Rebuilt fine-tuning workflows with Python , versioned data, and validated checkpoints for reproducible releases. Created verifiable rewards and RL environments for reasoning tasks, limiting reward-hacking risks. Profiled PyTorch and CUDA jobs, resolving multi-GPU stalls, memory fragmentation, and numerical instability . Partnered with edge engineers on quantization , balancing latency, context length, and model capability. Coordinated six engineers across evaluation, data, training, and runtime workstreams for agent research. Principal Machine Learning Engineer Redwood AI Systems January 2020 - June 2023 Denver, CO Led a post-training platform spanning ingestion, preprocessing, supervised fine tuning , reinforcement learning , checkpoint management, evaluation, and controlled model release. Balanced research speed with production quality while shaping deployment decisions through clear technical analysis for product, privacy, infrastructure, and hardware partners. Built post-training pipelines with tokenization and checkpoint controls, enabling dependable model releases. Converted research proposals into PyTorch trainers, evaluation adapters, and reward-model interfaces. Developed agent harnesses for context assembly, tools, skills , and persistent memory testing. Separated capability gains from prompt changes through factorial experiments and alignment regression suites. Traced 32-GPU failures across gradients, optimizers, data order, and collective communication. Presented model trade-offs to product, privacy, infrastructure, and hardware stakeholders during release planning. Senior Machine Learning Engineer Blue Mesa Research January 2017 - December 2019 Boulder, CO Directed language-model adaptation experiments focused on reasoning, instruction following, and tool-mediated completion within a research-to-production team. Mentored four engineers and established maintainable workflows that connected novel ideas with repeatable evaluation, operational readiness, and trustworthy model decisions. Implemented Python fine-tuning and policy optimization workflows with deterministic data preparation. Created factuality and alignment benchmarks combining automated scores with structured human review. Built RL environments with executable validators and shaped rewards for reasoning research. Optimized distributed jobs through batching, mixed precision, input pipelines, and GPU scheduling. Documented numerical failure modes, helping engineers diagnose training regressions during experiments. Mentored four engineers through design reviews, test coverage improvements, and operational planning. Machine Learning Engineer VectorForge Analytics July 2013 - December 2016 Pittsburgh, PA Developed predictive modeling and text-classification services from offline research through monitored production workflows. Built evaluation, lineage, and data-quality foundations that helped teams compare releases, investigate regressions, and turn ambiguous analytical questions into deployable services. Developed Python and scikit-learn services, moving predictive models into monitored production. Introduced dataset lineage and preprocessing checks, improving confidence in quarterly model comparisons. Built TensorFlow language-model prototypes with GPU acceleration and tested optimization settings. Created evaluation dashboards for accuracy, calibration, latency, drift, and error slices. Investigated scoring anomalies through data audits and replay tests, separating schema defects from regressions. Partnered with software and domain teams to convert questions into testable hypotheses. Software Engineer HarborPoint Software June 2010 - June 2013 Columbus, OH Built backend services and data-processing components in Python and Java for customer-facing applications. Established dependable APIs, automated testing, deployment workflows, and production diagnostics that formed a strong software foundation for later machine learning engineering work. Built Python and Java services supporting customer-facing applications across business units. Implemented REST APIs , batch workflows, database integrations, and automated release tests. Added structured logging and profiling, shortening diagnosis of long-running service failures. Automated Jenkins builds and Linux deployments, creating repeatable integration environments. Partnered with QA and product analysts on designs, acceptance criteria, and delivery plans. Standardized Python practices and interface contracts through code reviews and engineering documentation. PROJECTS Verifiable Reasoning Reward Lab 2024 Built a research sandbox for executable task validators, verifiable reward functions, and held-out reasoning tasks. Focused on trustworthy signals that help models improve without rewarding superficial shortcuts. Edge Agent Backend Study 2023 Compared compact language models under quantization, context, latency, and memory constraints. Connected runtime findings with agent tool use and practical on-device deployment decisions. Alignment Benchmark Observatory 2021 Created a benchmark analysis workspace covering capability, refusal behavior, tool-call validity, and human-reviewed alignment samples. Used repeatable comparisons to make model progress easier to discuss. LEADERSHIP & AWARDS Research Engineering Excellence Award, Redwood AI Systems, 2022 Best Applied Research Demonstration, Rocky Mountain Machine Learning Symposium, 2019 Graduate Fellowship in Applied Artificial Intelligence, University of Colorado Boulder, 2012 EDUCATION Master of Science in Computer Science University of Colorado Boulder GPA: 4.0 2012 Boulder, CO Coursework: Machine Learning, Natural Language Processing, Reinforcement Learning, Distributed Systems Bachelor of Science in Computer Science Ohio State University GPA: 4.0 2010 Columbus, OH Coursework: Algorithms, Statistics, Database Systems, Software Engineering CERTIFICATIONS AWS Certified Machine Learning - Specialty 2022 NVIDIA Deep Learning Institute: Building Transformer-Based Natural Language Processing Applications 2021 TensorFlow Developer Certificate 2020 TECHNICAL SKILLS Programming Languages: Python, Java, SQL Deep Learning Frameworks: PyTorch, TensorFlow, scikit-learn GPU Computing: CUDA, NVIDIA GPUs, mixed precision LLM Development: LLM architectures, transformer models, tokenization Post-Training: Supervised fine tuning, reinforcement learning, policy optimization Agent Systems: Agent harnesses, context engineering, agent memory Evaluation Systems: Evaluation frameworks, benchmarks, capability metrics Reward Engineering: Reward models, reward functions, verifiable rewards Data Engineering: Data ingestion, preprocessing, dataset versioning Distributed Training: Multi-GPU training, gradient scaling, collective communication Deployment Platforms: Model checkpointing, model deployment, cloud inference Edge AI: Quantization, edge inference, latency optimization MLOps Tools: Experiment tracking, CI/CD, Jenkins Operating Systems: Linux, shell scripting, structured logging Software Delivery: Git, REST APIs, automated testing SKILLS Python LLM architectures Model training Evaluation frameworks Benchmark design Agent harnesses Context engineering Agent memory Tool use Post-training pipelines Supervised fine tuning Reinforcement learning Reward modeling GPU optimization Technical leadership PROFESSIONAL AFFILIATIONS Technical mentor, Austin Machine Learning Engineering Forum, 2024 - Present Reviewer, Applied ML Systems Workshop, 2021 - Present Organizer, Colorado AI Research Practicum, 2018 - 2020 LANGUAGES English (Native) Spanish (Intermediate) ADDITIONAL INFORMATION Work Status : Authorized to work in United States. No sponsorship required. ADDITIONAL INFORMATION Work Status : Authorized to work in United States. No sponsorship required. REFERENCES AVAILABLE ON REQUEST

File
- Open
- Save
- Save As
- Rename
- PDF Page Setup
- Preview
- PDF Download
- Show Styles
Preview
Show Styles
PDF Download
US English
US English
UK English
Spanish (LatAm)
French
German
Japanese
Italian
Portuguese (BR)
Chinese (Simplified)
Log In