Role Description :

We are hiring a Senior AI/ML Architect Agentic Conversational AI (6+ Years Experience | 100% Remote) to own and evolve the end-to-end AI architecture behind Hylys conversational platform.

This is a hands-on architectural leadership role for someone with 6+ years of production AI/ML experience, including deep experience in :

- LLM orchestration & agentic workflows

- Retrieval-Augmented Generation (RAG) Real-time, low-latency serving

- Evaluation-driven development

- Production observability and rollout safety

You will define how AI systems are designed, measured, deployed, and scaled across multiple channels.

Must Have Qualifications :

- 6+ years of professional AI/ML systems experience

- 3+ years building conversational AI systems in production

- Proven experience deploying LLM-based architectures (OpenAI, Claude, Llama/DeepSeek-class)

- Hands-on experience with agentic systems :

a. Tool calling

b. Structured outputs

c. Policy control

- Strong RAG expertise with vector databases and reranking

- Deep understanding of latency and cost optimization

- Strong Python skills : experience with async services (Fast API preferred)

- MLOps experience :

a. CI/CD

b. Model & prompt versioning

c. Containers and Kubernetes

d. Feature flags and canary releases

- Strong systems design and cross-team leadership skills

- Experience with observability in production AI systems

Job Responsibilities :

Architecture & Platform Ownership :

- Own the architecture of Hylys agentic conversational platform across chat, voice, SMS, and email

- Define service boundaries, interface contracts, and scalability patterns

- Ensure reliability, modularity, and extensibility

Agentic Orchestration :

- Design and evolve the orchestrator layer :

a. Intent routing

b. Tool schemas

c. Policy enforcement

d. Memory management

e. Structured outputs

RAG & Knowledge Systems :

- Build and standardize RAG pipelines :

- Hybrid retrieval

- Embeddings + reranking

- Grounded responses with citations

Performance & Cost Optimization :

- Optimize real-time inference using :

- Keyword & semantic caching

- Streaming, batching, prompt compression

- Cost/latency trade-off analysis

Evaluation & Quality :

- Establish eval-driven delivery :

a. Offline and online benchmarks

b. Regression detection

c. Rollout gates tied to business metrics

- Ensure AI quality improves continuouslynot just features

Observability & Reliability :

- Implement production-grade observability :

a. Distributed tracing

b. Error taxonomies

c. SLOs and alerting

- Lead incident response and postmortems

Leadership & Standards :

- Mentor engineers and set AI architecture standards

- Define best practices for versioning, rollouts, and documentation

- Partner closely with Product, API, FE, and QA teams

Nice to Have (Qualifications) :

- Remote Team Management Experience : Experience managing teams in a remote environment, ensuring effective communication and productivity

- Cross-Functional Team Leadership : Proven ability to lead cross-functional teams, managing dependencies and ensuring smooth coordination

- Strong collaborative development skills using Git and version control systems