We're looking for a Forward Deployed AI Engineer who combines strong software engineering fundamentals with hands-on AI/LLM engineering and a customer-first mindset. You'll embed with customers and internal partner teams to design, build, and ship production-grade, AI-powered solutions - from intelligent integrations to LLM-driven agents and services - translating real-world business problems into robust, scalable systems.
This is a role for engineers who love writing strong Java code, building with modern AI, and working shoulder-to-shoulder with the people using what they build. You won't be handed a spec and left alone: you'll uncover requirements, make architectural calls, wire up AI capabilities that actually work in production, and own the outcome end to end.
Location: Remote in LATAM. Working hours are based on the US Central or Eastern Time Zone.
About the Company:
Abstra is a fast-growing, Nearshore Tech Talent services company, providing top Latin American tech talent to U.S. companies and beyond. Founded by U.S.-bred engineers with over 15 years of experience, Abstra specializes in sourcing skilled professionals across a wide range of technologies to meet our clients’ needs, driving innovation and efficiency.
What You'll Do
- Build production services in Java — design and implement microservices (Spring Boot) that integrate customer systems with our platform and AI capabilities.
- Engineer AI-powered features — build LLM/GenAI applications using model APIs (Anthropic Claude, OpenAI, AWS Bedrock, etc.): RAG pipelines, agentic workflows, tool/function calling, prompt engineering, and evaluation.
- Deploy and operate on the cloud — build, ship, and run services and AI workloads on AWS, containerized and orchestrated with Kubernetes.
- Embed with customers/partners — work directly with client teams to gather requirements, design AI solutions, and drive them to go-live — including responsibly setting expectations about what AI can and can't do.
- Own the full lifecycle — discovery, design, implementation, evaluation, deployment, and post-launch support of both services and AI features.
- Make AI production-ready — handle the hard parts: grounding/hallucination control, latency and cost optimization, guardrails, observability, and evaluation/testing of non-deterministic systems.
- Debug across the stack — diagnose issues spanning distributed systems, APIs, data pipelines, model integrations, and infrastructure.
- Translate ambiguity into architecture — turn loosely-defined business needs into clear technical designs; push back when there's a better or safer approach.
- Improve the platform — feed field learnings back into the product; build reusable AI patterns, tooling, prompts, and documentation.
What You'll Need (Required)
- 3+ years of professional software engineering experience with strong Java (Java 8–21).
- Hands-on experience with Spring Boot and building/consuming RESTful APIs.
- Applied AI/LLM engineering experience — you've built and shipped something real with LLMs: e.g. RAG, agents, tool calling, prompt engineering, or model-API integration (Claude, OpenAI, Bedrock, Gemini, or similar).
- Exposure to AWS — deploying and running applications using core services (EC2, S3, IAM, RDS, Lambda, CloudWatch); familiarity with AWS AI/ML services (e.g. Bedrock, SageMaker) a strong plus.
- Exposure to Kubernetes — deploying, running, and troubleshooting containerized workloads (Docker + K8s).
- Solid grasp of relational databases (SQL); familiarity with vector databases / embeddings for retrieval.
- Strong debugging and problem-solving skills across distributed and AI-integrated systems.
- Excellent communication — comfortable working directly with customers and non-technical stakeholders, including explaining AI capabilities and limitations.
- Bachelor's degree in Computer Science or equivalent practical experience.
Nice to Have (Preferred)
- Experience with Kotlin or Python (common for AI/ML tooling); JVM build tools (Maven / Gradle).
- Agentic frameworks / orchestration — LangChain, LlamaIndex, Spring AI, Model Context Protocol (MCP), or similar.
- LLM evaluation & observability — building eval harnesses, prompt/version management, tracing (LangSmith, Langfuse, or homegrown).
- Classic ML / MLOps — model training, fine-tuning, feature stores, model serving, and deployment pipelines.
- CI/CD and infrastructure-as-code (Terraform, Helm, CloudFormation).
- Event-driven / streaming systems (Kafka, SQS/SNS) and microservices patterns.
- Awareness of responsible AI — safety, guardrails, PII handling, prompt-injection defense, and data privacy.
- Prior customer-facing / consulting / implementation engineering experience.
What We Offer:
- Flexible working hours and remote work options.
- Opportunities for professional growth and development.
- A collaborative and inclusive work environment.
- The chance to work on impactful projects with a talented team.
- Excellent compensation in USD.
- Hardware and software setup.
