AI Engineer
About the job
- At least 5 years of professional software engineering experience, including at least 1 year shipping LLM-based or AI-powered features to production.
- Hands-on experience designing or building agentic systems, such as tool calling, multi-step reasoning, planning loops, or agent orchestration.
- Working knowledge of the Model Context Protocol (MCP), or demonstrated ability to learn it quickly, plus familiarity with agent and tool standards.
- Strong Python skills and experience building and deploying backend services and APIs using FastAPI, Flask, or similar technologies.
- Hands-on experience with at least one major LLM provider, including OpenAI, Anthropic, Bedrock, Azure OpenAI, Vertex, or open-weight models via vLLM or Ollama.
- Working knowledge of retrieval-augmented generation (RAG), including embeddings, vector databases such as pgvector, Pinecone, Weaviate, or Qdrant, and retrieval evaluation.
- Comfort with prompt engineering, structured outputs such as JSON mode and schemas, and tool or function calling.
- Experience writing evaluations for non-deterministic systems.
- Solid SQL skills and experience with relational and unstructured data.
- Familiarity with at least one cloud platform: AWS, Azure, or GCP.
- Experience with Git, code review, and modern collaborative workflows.
- Strong written and verbal communication, including the ability to explain AI tradeoffs to non-technical stakeholders.
- Must be a U.S. citizen or legal resident and able to work domestically.
- Must be able to attain a low-level security clearance.
- Must work from the United States.
- Design and build agentic systems, including multi-step agents that plan, call tools, retrieve context, and take action with appropriate human-in-the-loop checkpoints.
- Build MCP servers and clients to securely expose client data, internal tools, and APIs to large language models in a standardized, auditable way.
- Ship large-language-model-powered applications such as copilots, document intelligence, search, summarization, and workflow automation.
- Design and maintain RAG pipelines covering chunking, embeddings, vector stores, retrieval, reranking, and grounding.
- Integrate model APIs from OpenAI, Anthropic, Bedrock, Azure OpenAI, and open-weight models; select models based on quality, latency, and cost.
- Develop evaluations and observability for agents and AI features to monitor production performance and regressions.
- Apply prompt engineering, structured outputs, function and tool calling, and guardrails to make agent behavior predictable.
- Write production Python backends and APIs that expose AI capabilities to web and mobile clients.
- Collaborate with engineers, designers, and product staff to scope appropriate AI functionality for products.
- Help shape responsible AI practices for federal use, including privacy, security, auditability, and human oversight.
- Experience authoring MCP servers for non-trivial systems such as databases, internal APIs, and document stores.
- Experience with evaluation and observability platforms such as Braintrust, LangSmith, Langfuse, Arize, or custom harnesses.
- Experience with multi-agent orchestration patterns and reasoning about agent failure modes.
- Fine-tuning, distillation, or LoRA experience that has produced meaningful results.
- Experience with Docker, Kubernetes, and CI/CD for AI workloads.
- TypeScript or Node.js experience for full-stack AI features.
- Experience with streaming user interfaces using SSE or WebSockets and token-level user experience patterns.
- Experience with caching, prompt compression, and cost or latency optimization at scale.
- Background supporting federal or government clients.
- Awareness of the NIST AI Risk Management Framework, FedRAMP, or related responsible-AI frameworks.
About the company
Onvista provides a financial portal that gives private investors access to real-time stock quotes, market data, and financial news. Users interact with the platform by tracking securities through personalized portfolio management tools and using charting features to analyze company performance. Unlike many standalone news sites, Onvista integrates these information services directly with its own online brokerage, allowing users to move from research to trading in one place. The company aims to provide German-speaking investors with all the data and execution tools necessary to manage their own investments independently.
Company Size
11-50
Company Stage
N/A
Total Funding
N/A
Headquarters
Washington DC, District of Columbia
Founded
2004
Get referred to BLEN
See people who can refer or advise you
Simplify's Take
What believers are saying
- Q2 2026 revenue reached $33.8 million, up 7%, near the guidance high-end.
- Autopilot has over 65 lenders active and six signed contracts by August 2026.
- Frost Bank lifted completion rates 60%, proving measurable ROI for lender customers.
What critics are saying
- Mortgage rates above 6.8% in August 2026 keep refinancing weak and compress growth.
- Blend lost HMDA share to 17% in 2025, reflecting customer churn and roll-offs.
- ICE Mortgage Technology and nCino consolidate lenders onto bundled stacks, threatening Blend's standalone POS role.
What makes BLEN unique
- Autopilot completed loan reviews in fifteen seconds; commercial launch began July 1, 2026.
- Blend is OpenAI Select Partner since August 27, 2026, strengthening enterprise AI credibility.
- Its platform spans mortgage, consumer banking, deposits, and MCP-enabled lender workflows.
Help us improve and share your feedback! Did you find this helpful?
Benefits
Remote Work Options
Health Insurance
Company News
BLEN named an OpenAI Select Partner. * BLEN Team BLEN Corp is excited to share that BLEN has been named an OpenAI Select Partner and is now part of the OpenAI Partner Network. For a company that has spent more than two decades building software for federal agencies and mission-driven organizations, this is a meaningful milestone. It reflects the work its team has already put into shipping AI solutions in production, and it gives BLEN Corp a stronger footing to help its customers build, deploy, and scale AI responsibly. What the partnership means. The OpenAI Partner Network brings together organizations with industry expertise and delivery experience to build, sell, and deliver AI solutions with OpenAI. As an OpenAI Select Partner, BLEN will continue working with OpenAI to help organizations build, deploy, and scale AI responsibly and effectively. For its customers, its focus remains: * Moving from pilot to production. BLEN Corp use its delivery playbooks to move beyond proofs of concept and build systems that real users depend on. * Informed architecture decisions. BLEN Corp evaluate models, context management, and tool use against each customer's requirements, testing reliability and cost before deployment. * Responsible deployment by default. Security, privacy, accessibility, and compliance are not afterthoughts in government work. Its approach to AI carries the same standards BLEN Corp apply to everything else BLEN Corp build. Why this matters for federal and enterprise customers. Most of its customers are not asking whether AI is interesting. They are asking how to use it safely inside environments with strict data handling rules, legacy systems, and real accountability to the public. That is the problem space where BLEN has always worked. Its Intelligent Automation practice already combines process automation, machine learning, and large language models to reduce manual effort and improve decision-making. Recent work includes AI-enabled clinical summarization and Model Context Protocol (MCP) server support for the Department of Veterans Affairs, where the bar for correctness and trust could not be higher. Joining the OpenAI Partner Network strengthens that practice. What's next. Looking ahead, BLEN Corp plan to expand its work in: * Summarizing and structuring large volumes of documents and case records * Building agents that connect models to internal systems through MCP and well-defined tools * Modernizing legacy workflows with AI-assisted intake, triage, and routing * Standing up evaluation and monitoring so AI features stay reliable after launch BLEN Corp is proud to be part of this next chapter. Let's talk. If you're a federal agency or mission-driven organization figuring out how to put AI to work safely, BLEN Corp'd like to hear about it. Whether you have a pilot that needs to reach production or a workflow that's overdue for modernization, get in touch or email BLEN Corp at [email protected].