AI Engineer @ lunartree.bio

About Lunartree

LunarTree is building an AI-native platform that is rewiring how new medicines are made - by giving leaders working on the next generation of drugs, the AI intelligence layer they always wish existed.

Our flagship offering, Biolens, helps some of the biggest companies in the world save hundreds of hours by eliminating painful busywork and helping teams focus on what matters — making the best decision quickly.

We've got incredible backers from Pfizer, Amgen, Merck, BCG, A91 Partners behind us.

Our team includes former founders, investors, consultants, athletes, and engineers from IIT Bombay, Goldman Sachs, Merkle, SuperDM.

We're on a mission to eliminate busywork in bringing novel treatments to the world.

And now we need someone to make sure our infrastructure doesn't break as we scale.

About the Role

• We're looking for a sharp, hungry engineer who will run all the AI experiments we want to try but don't have time for.

• You'll work directly with the founder (Mukesh) to explore new models, improve quality, optimize costs, and figure out what's possible with LLMs and agents.

• This isn't about shipping features on a roadmap - it's about rapid experimentation to find what works and what doesn't. You'll try things, measure results, report findings, and move fast.

• High autonomy, cutting-edge AI work, and direct impact on product quality.

What You'll Do

• Try new models: Test Claude, GPT, Gemini, open-source models - figure out which performs best for our use cases and at what cost.

• Improve quality: Experiment with prompting strategies, few-shot learning, RAG techniques, reasoning improvements, guardrails.

• Explore RAG: Test chunking strategies, retrieval methods, embedding models, vector databases - optimize for accuracy and speed.

• Build agent capabilities: Work on memory management, tool design, multi-step reasoning, context handling.

• Optimize systems: Run cost vs quality tradeoffs, measure latency, find ways to get better results cheaper and faster.

• Research and implement: Read papers, try new techniques, adapt them to our pharma/biotech domain.

Essentially: "here's an idea - go figure it out and come back with findings"

Who You Are

• Strong fundamentals: CS degree from IIT/BITS (or equivalent) with solid grasp of algorithms, data structures, systems thinking.

• Self-learner: Comfortable diving into new frameworks, tools, papers. Can figure things out from docs and experimentation.

• Proactive: Don't need hand-holding. You see what needs testing and just do it.

• Excited about AI: You want to push the boundaries of what LLMs and agents can do, not just use them.

• Scrappy experimenter: You move fast, measure results, document learnings, and iterate.

Experience Level

Fresh grad or 1-2 years experience. We care more about raw ability and hunger than years of experience.

Bonus if

• Previous work with LLMs, RAG systems, or agent frameworks

• Published research or strong GitHub projects in AI/ML

• Experience with prompt engineering or model fine-tuning

• Contributed to open-source AI projects

• Familiar with life sciences domain (or excited to learn it)

Not a Fit If

• You want well-defined tasks and clear requirements (this is exploratory work)

• You need a lot of structure and guidance (you'll be figuring things out independently)

• You're not comfortable with things not working (most experiments fail, that's the point)

Tech Stack You'll Work With

Python, LangChain, LlamaIndex, Claude/GPT/Gemini APIs, vector databases, PostgreSQL, AWS, various AI frameworks and tools

Location

Hybrid (Bangalore)

Compensation

18-24 LPA + ESOPs

Hiring Process

Apply below - we reply within 72 hours.

One technical exercise (AI/systems problem).

In-person/virtual interviews with the founders.