Senior AI Engineer — Inference & Agent Systems

Arcana; Bengaluru; Hybrid; INR 4,000,000 - 7,000,000; Full time

Job summary

Arcana is building real-time AI agents that synthesize information across sources, delivering structured, reasoned answers. The Senior AI Engineer role focuses on optimizing inference pipelines, streaming responses, and robust orchestration atop Temporal, with a strong emphasis on latency, reliability, and end-to-end eval Harnesses.

You’ll own Plan-Execute-Synthesize pipelines, ensure robust tooling for tool calls, and drive latency regression testing across deployments in a fast-moving team.

Qualifications

• :You're a signal of practical production experience at meaningful scale.
• :
• You understand why latency matters and how to measure it.
• You have built multi-step agent systems and systems that tolerate non-determinism.

Responsibilities

• Drive TTFT below 400ms for multi-step agent pipelines
• Streaming optimization: deliver first token while sub-agents run
• Design and implement Plan-Execute-Synthesize pipelines with parallel sub-agents
• Build reliable orchestration on Temporal: retries, timeouts, idempotency
• Implement JSON schema validation and graceful degradation
• Own eval framework from data to scoring and regression detection
• Develop tooling to route across providers by latency and cost

Skills

TTFT optimization

Streaming latency

Latency budgets

LLM orchestration

Observability

Tool call design

Production systems

Evaluation harness

Tools

Go

Python

Temporal

Kafka

PostgreSQL

Docker

Job description

Senior AI Engineer — Inference & Agent Systems

Title: Senior AI Engineer — Inference & Agent Systems

Location: - BLR/Remote India

What We’re Building

Arcana is building AI agents that synthesize information across heterogeneous sources and deliver structured, reasoned answers in real time. The product only works if the agents are fast, reliable, and correct, not approximately correct.

Our stack: Go + Temporal for orchestration, a Plan-Execute-Synthesize agent architecture, and an evaluation harness we use to measure every regression. The problems are hard. The latency bar is aggressive. The accuracy requirements are unforgiving.

The Work

Inference Optimization

- Drive TTFT below 400ms for multi-step agent pipelines

- Streaming optimization: first token to user while sub-agents are still running

- KV cache strategy, prompt compression, dynamic context window management

- Multi-provider routing: model selection by latency, cost, and task type across OpenAI, Anthropic, Gemini, and open-weight models

- Design and implement Plan-Execute-Synthesize pipelines that run sub-agents in parallel DAGs, not sequential chains

- Build reliable orchestration on top of Temporal: retries, timeouts, partial failure recovery, idempotency

- Structured output enforcement: JSON schema validation, retry loops on malformed LLM output, graceful degradation

- Tool call design: schema design that LLMs actually follow reliably across providers

- Own the eval framework end to end: ground truth datasets, automated scoring pipelines, regression detection on every PR

- LLM-as-judge pipelines for qualitative output assessment

- Latency regression testing - p50/p95/p99 tracked across every deployment

- Adversarial test case design: ambiguous queries, missing data, conflicting sources, malformed tool responses

Infrastructure

- Model serving and cold start optimization

- Async worker architecture for parallel sub-agent execution

- Observability: trace every token, every tool call, every synthesis step

What We’re Looking For

You've built something that runs in production at a meaningful scale and you understand why it's fast (or why it isn't).

Strong signal:

- You've worked on inference pipelines where TTFT was the primary metric and you moved it meaningfully

- You've built multi-step agent systems and you know where they break not from reading papers but from watching them fail in production

- You've written eval harnesses from scratch and you have opinions about what makes a ground truth dataset actually useful

- You've debugged LLM non-determinism in production and built systems resilient to it

- You've worked with streaming LLM responses and built infrastructure around partial output handling

Weaker signal (but not disqualifying):

- You've fine-tuned models but haven't shipped inference systems

- You've used LangChain/LlamaIndex but haven't built the layer underneath

- Strong ML research background without systems exposure

Stack familiarity (we care more about depth than match): Go, Python, Temporal, Kafka, PostgreSQL, Docker

Why This Role

The problems here don't have blog posts about them yet. Parallel agent DAG execution under hard latency budgets, streaming synthesis across partial sub-agent results, eval harnesses for non-deterministic multi-step systems: these are genuinely unsolved at production quality. Small team. High ownership. Every engineer’s decisions ship to production.

Who We Want to Hear From

You've shipped inference systems at:

- A real-time AI product (search, coding assistant, chat at scale)

- An agent platform (any domain)

Or you've built eval/harness infrastructure that a team of 10+ engineers actually trusted to catch regressions.